This PDF file contains the front matter associated with SPIE Proceedings Volume 13445, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

The goal of this study is to apply machine learning methods in low-frequency time code multi-path signal separation. Through a comparative analysis, traditional signal separation techniques like adaptive filtering and Independent Component Analysis (ICA) are weighed against three notable machine learning algorithms: Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM). Through simulation experiments, the feasibility of machine learning in low-frequency time code signal separation was verified. The results affirm the superiority of machine learning algorithms, with LSTM notably showcasing the highest output signal-to-noise ratio and minimized mean squared error. Additionally, CNN excels in scenarios with smaller signal delays and suboptimal signal-to-noise ratios, while RNN performs between CNN and LSTM. Furthermore, for similar one- dimensional signal separation problems, machine learning, especially LSTM, is also a good solution.

This paper summarizes the key application and importance of DL (Deep Learning) technology in communication system performance optimization. First of all, it emphasizes the key role of communication system in modern society and economy, and the challenges brought by the increasing communication demand. In addition, the multiple applications of DL in communication systems are emphasized, including signal processing and beamforming, channel estimation and equalization, resource allocation and power control. DL model can also be used to detect network intrusion and malicious attacks. In this experiment, DL model is compared with traditional methods in network intrusion detection rate. At different time points, the detection rate of DL model fluctuated, but remained at a high level on the whole. At 12 hours, the detection rate is close to 0.9. Compared with DL model, the detection rate of traditional method fluctuates greatly, and it is obviously lower than DL model at some time points. At the time point of 12 hours, the detection rate of traditional methods is about 0.7. It can be seen that DL model can effectively detect network intrusion and malicious attacks at different time points and has better performance than traditional methods.

In order to improve the intelligent control effect of building electrical, this paper combines wireless communication technology with Internet of Things technology and proposes a new type of building intelligent system based on Internet of Things technology, which has the characteristics of no center, self-organization and flattening. The building electrical equipment under this system building can realize wireless self-organization network, plug and play, and coordinated operation. This equipment management information system refers to the equipment life cycle management concept of equipment management in Industry 4.0 era, and manages the information generated by all activities in the whole life cycle of equipment. Moreover, the operation data of four kinds of electrical equipment commonly used in office buildings are selected for analysis, and the probability conversion process of users' actions between startup and shutdown.

With the advent of the information age, the power industry, as an important pillar of the national economy, is facing dual challenges of market expansion and diversified user demands. In order to improve the operational efficiency and service quality of the power system, ensure the stability and reliability of power supply, this article focuses on the design and implementation of a telephone channel operation and control system. The system interacts with users in real-time through telephone channels, quickly responding to inquiries and repair needs, and monitoring and analyzing the operating status of the power system in real-time, providing technical support for the stable operation of the power system. This article provides a detailed description of the system's functional and performance requirements and adopts a modular design approach to ensure the flexibility and scalability of the system. This article demonstrates significant improvements in key performance indicators of the new system, including peak traffic processing capacity, average connection time, customer satisfaction, and work order processing time, thereby demonstrating the significant advantages of the new system in improving service efficiency, customer satisfaction, and reducing operating costs.

This paper introduces an intelligent flowerpot system, which aims to solve the problem of flower death caused by busy or little understanding of flower habits. The system is based on the Internet of Things and WeChat applets, and uses sensors such as temperature and humidity, light and soil moisture to obtain information about the growth environment of flowers and realizes automatic watering and light control. The system is based on the Internet of Things and WeChat applets, and uses sensors such as temperature and humidity, light and soil moisture to obtain information about the growth environment of flowers and realizes automatic watering and light control. uses the Alibaba Cloud IOT platform to transmit sensor data to the WeChat Mini Program, through which users can monitor and control the flowerpot system in real time and understand the status of flowers. At the same time, the system uses the Alibaba Cloud IOT platform to transmit sensor data to the WeChat Mini Program, through which users can monitor and control the flowerpot system in real time and understand the status of flowers. This kind of intelligent flowerpot system realizes the interactive communication between flower growers and flowerpots, which is of This kind of intelligent flowerpot system realizes the interactive communication between flower growers and flowerpots, which is of practical significance for enhancing the planting skills and experience of flower raisers.

Brushless DC motor has high reliability and strong anti-interference ability under the addition of position sensor technology, and position sensor-less control has become one of the hot spots in the research of brushless DC motor. In this paper, the second-order super-helical sliding mode observer is designed for the shortcoming of the traditional sliding mode observer, which is easy to cause large errors in the calculation of the arctangent function, replacing the traditional arctangent function with a phase-locked loop, and replacing the traditional symbolic function with a new type of continuous function. By building a simulation model and verifying it, the method used in this article can not only reduce the impact of vibration, but also improve observation accuracy and reduce the error in observation.

During the power inspection process, image and video data are generated, and building visual models based on these data for defect recognition and fault diagnosis has been widely applied in the power industry. Due to external factors such as weather and transportation, adversarial samples have become an important factor in reducing the efficiency of visual models. In response to this issue, a credibility evaluation method for power vision models is proposed in this paper. Analyze the mechanism of adversarial sample generation in image and video data generated during power inspection and extract common features of adversarial samples through cross sample comparison and matching. Then, based on the accuracy and efficiency requirements of visual models for power application scenarios such as transmission channel monitoring, drone inspection, and safety monitoring, design visual model efficiency and credibility evaluation rules. Building a visual model credibility evaluation model based on the efficiency and credibility evaluation rules of electric power visual models, to calculate the accuracy, precision, recall of visual object classification tasks, as well as the average accuracy and recall of detection tasks. Calculate the interference intensity of adversarial samples based on the weight calculation results, form the corresponding credibility evaluation report of the power vision model, and feedback it to the data management party, thereby achieving closed-loop management of the basic data quality of the power vision model, improving the sample classification effect and recognition accuracy of the power vision model.

Firstly, this article introduces the background and purpose of Low-Voltage Distribution Network (LVDN) topology intelligent identification system. The characteristics and importance of LVDN are expounded. The method and steps of applying machine learning (ML) algorithm to intelligent topology identification system of LVDN are introduced in detail. Then, the relevant data of LVDN are collected, preprocessed and feature extracted. The intelligent topology identification model of LVDN is constructed by using Deep Learning (DL) algorithm, and it is trained and optimized. Finally, this article makes a simulation study on the topology intelligent identification system of LVDN. A LVDN simulation model was established, corresponding topology structures were generated, and simulation data was input into the LVDN topology intelligent recognition model for training and testing; The identification results were analyzed and compared, verifying the feasibility and effectiveness of this method. By comparing the experimental results with the performance of traditional methods, it is found that the efficiency of this system is improved by about 19% compared with traditional RNN. Moreover, the recognition accuracy of the system is high, which can reach about 96%. Its research has certain theoretical and practical value.

In microgrids, sag control strategies are commonly employed by inverters across various levels. This study introduces an innovative solution to address challenges in sag control, specifically targeting issues related to uneven reactive power distribution and present transmission line impedance causes the AC bus voltage to drop. The proposed approach involves the implementation of an adaptive virtual impedance method and a feedback bus voltage recovery strategy, both grounded in the principle of multibody consistency. Firstly, the paper analyzes the causes of uneven reactive power distribution and busbar voltage deviation caused by traditional inverter sag control with multi-inverters parallel model. Secondly, the improvement strategy of sag control is proposed, and the virtual impedance and bus voltage feedback method of control based on distributed consistency principle is introduced to achieve the purpose of reactive power equalizing and restoring bus voltage. Finally, two inverter models were built in the MATLAB/Simulink, validation of the proposed strategy and demonstrate the effectiveness of the proposed strategy. Simulation results show that the proposed strategy can effectively improve the performance of the system of rational reactive power distribution and stable bus voltage of multiple inverters and can provide reference for the parallel use of multiple inverters

The industrialization of various countries has promoted the continuous consumption of energy, which has led to increasing demand, and shale oil has been gradually explored and utilized by the public as a substitute for oil. As the mainstream technology in oil shale mining, in-situ mining heating technology is widely used. A new oil shale development technique is introduced in response to the low energy utilization rate of in-situ mining heating technology. Oil shale's tiny, carbonized channels are immediately penetrated by voltage in the power heating technology, which then heats the oil shale to pyrolyze it. This article uses simulation software to establish a coupled model, obtains the temperature distribution and potential distribution of oil shale through numerical calculation, and analyzes and compares it with traditional in-situ mining techniques. According to the findings, power supply heating has a 32.4% greater heating rate than traditional heating, a 27-fold increase in the effective heating volume, a faster heating rate, and a higher energy utilization rate.

For the purpose of enhance the steady-state accuracy and rapid reaction of the direct current unilateral bus bar voltage and actual export power of the compressed air energy storage expansion power generation structure, a modified linear active disturbance rejection control technique account of lumped disturbance observation error is created and used for voltage outer loop to take shape a closed loop construction that has never been created and used. Based on this the design theory of linear active disturbance rejection control means is illustrated. Meanwhile, the simulation model is constructed to testify it. The conventional linear active disturbance rejection control means as well as PI control means are respectively contrast with the means mentioned in this article. The simulation conclusion indicates a phenomenon which about modified linear active disturbance rejection control is better than mentioned above two conventional methods, which can make the actual export power of the grid side more quickly track the required power and remain stable, meanwhile, the stable performance of the direct current bus voltage is also improved.

The digital twin model of distribution network has been widely used in power maintenance and equipment status monitoring, but the data update speed of the digital twin model is limited by technologies such as data collection and data management, and the real-time analysis ability of the digital twin model is insufficient. In response to this issue, this paper proposes a high-performance digital twin analysis technique for distribution networks based on parallel acceleration. Analyze the application scenarios of the digital twin model in the distribution network and identify the data request characteristics of the digital twin model. Classify and identify model data resources based on the data request features in the digital twin model and construct corresponding data storage carriers based on the classification identification results. The same type of data request access only requires extracting data from a fixed data storage carrier, and various types of data call requests do not interfere with each other, thus achieving parallel data calls. Then upgrade different data storage carriers, construct corresponding memory areas for each data storage space, and improve the input and output performance of each data storage carrier. Based on this, the optimized digital twin model will be applied to the maintenance of power equipment, integrating historical and real-time data of power equipment maintenance, conducting analysis and application of power equipment maintenance, and verifying the effectiveness of the method.

In order to improve the accuracy of diagnosis, a fault diagnosis and research method of gearbox bearing of wind turbine based on ICEEMDAN-PE and GWO-LSSVM is proposed. The method firstly uses ICEEMDAN to decompose the sampled data signals, then uses PE to extract the features, and finally inputs the features into Grey Wolf algorithm (GWO) to optimize the Least Squares Support Vector Machine (LSSVM) diagnostic model to train the optimal parameters. A fault diagnosis model is built to identify common faults in gearbox bearings accurately. Combined with the simulation experimental data, the diagnosis accuracy is 97.67%, which proves the feasibility of the scheme, and provides an effective method for improving the accuracy of the gearbox bearing fault diagnosis

In the background of the increasing shortage of fossil energy, electric vehicles gradually enter people's lives, and how to achieve efficient feedback braking of electric vehicles has become the focus of people. In this regard, based on the principle of electric vehicle feedback braking, in this paper, a simulation model is established, which combines the characteristics of vehicle braking process and uses particle swarm optimization algorithm to optimize the trolley feedback system. The simulation results show that the optimized scheme can guarantee the braking stability of the vehicle.

The artificial intelligence big model has been successfully applied in the maintenance of power equipment, but power equipment maintenance involves multiple modalities of data. The high computational complexity and storage requirements make it difficult to widely promote under limited conditions. In response to this issue, a model compression method based on knowledge distillation is proposed in the paper to simplify the artificial intelligence cognitive large model for power equipment operation and inspection. The operation and inspection data of power equipment generally have complex data or knowledge backgrounds, and the target knowledge in the knowledge base is relatively small. To avoid data noise, the updated data from the previous iteration is used to guide the training of the model for the next iteration. Separate the feature information of the teacher model into local and global features and train the student model separately by calculating local and global losses. Introducing knowledge adjustment mechanisms and dynamic temperature distillation mechanisms, allowing student models to autonomously select the core features of images, in order to reduce the misleading effects of incorrect predictions in the model. Finally, the application is combined with the operation and inspection data of power equipment to verify the effectiveness of the model lightweight method described in the paper.

This paper investigates the intelligent control algorithms and applications for mechatronic systems of robots. The fundamental principles and components of robot control in mechatronic systems are introduced, including perception, decision-making, and execution. Sensors, controllers, and actuators are highlighted as core components. Subsequently, intelligent control algorithms for robots are discussed in detail, incorporating intelligent algorithms such as reinforcement learning, deep learning, fuzzy control, and evolutionary algorithms to enhance the robot's perception, decision-making, and execution capabilities. These algorithms possess adaptability, learning capability, and optimization characteristics, allowing robots to make intelligent decisions and achieve precise execution in complex environments. Finally, the applications of intelligent control algorithms for robots in mechatronic systems are summarized. In automated production lines, they contribute to the intelligence and flexibility of the production process. In the field of service robots, they facilitate human-robot interaction and service delivery. In medical robots, they enable surgical assistance and rehabilitation training, among other functions. Through the research and application of intelligent control algorithms for robots, the performance of robots can be effectively improved, expanding their application areas, and driving the development of robot technology in mechatronic systems. This study carries significant theoretical and practical implications for robot control and application.

As the power system rapidly develops, ensuring stable operation of substations enhances the reliability of the power grid. Traditional fault detection methods of substation equipment depend on manual inspection and regularly scheduled maintenance, which is inefficient and cannot be conducted in real time. In recent years, computer vision technology has rapidly advanced, providing new solutions for the automatic detection of substation equipment. It looks at, considering deep learning theory, the application of computer vision technology to the automatic detection of substation equipment; more precisely, through the use of CNNs, R-CNNs, and GANs, the performance relating to fault detection and classification. Through the deep learning model construction and feature extraction as well as fault classification design, test the effectiveness of the model. Thus, this research improves not only accuracy and efficiency in fault detection but also offers technical support for automation and intelligent management of substations. Research results: Deep learning technology has enormous application prospects in fault diagnosis of substation equipment. However, restricting issues in data collection, resources required for model training, and technical solution economic costs still need further study.

In this paper, the application of robot technology to welding engineering in automotive industry is studied. Based on the research and analysis of the relationship between the robot body and its external equipment and the actual production, through the process of hardware integration and path planning, the fixture system and motion command modules of the robot integration platform are developed, and the design and construction of the whole robot system platform is completed. The feasibility of the scheme is verified by actual project research and development. Practice has proved that it is feasible to apply the robot system platform to the automotive welding engineering process, which can improve the production efficiency and management efficiency, improve the welding precision and production stability. The method and practice of robot system integration in this paper can be used for the integration of general industrial process systems of general robots and can provide reference for the development and construction of industrial robot systems.

This paper studies and realizes a low-cost stroke control algorithm for agricultural push rod motor. The algorithm is based on the hall counting principle. By adding two hall sensors to the ordinary agricultural push rod motor, the motor can output two square wave signals when it is running. When the motor is running, each revolution will trigger a hall sensor to output a pulse signal. The two signals have a phase difference of 90 degrees. Through the phase relationship of the two signals, the forward and reverse state of the motor can be judged, and the pulse signal can be counted by adding or subtracting. The algorithm realizes the automatic control of motor stroke according to the user set stroke percentage, hall count value and total stroke hall count value. This algorithm has the advantages of stable and reliable stroke control, low cost, suitable for high humidity and dusty environment in agriculture, and can be applied to the on-site automatic control of related equipment in facility agriculture and field agriculture, as well as remote precision control based on Agricultural Internet of things.

This paper presents a detailed study of the design, simulation, and analysis of a CMOS image sensor, focusing on a 3T active pixel sensor (APS) readout circuit and a 3x3 pixel array. Utilizing SPICE simulations, this paper meticulously investigates the sensor's response to varying light intensities and assesses its performance in capturing high-resolution, low-noise images. The research introduces a novel approach for image reconstruction from voltage signals, employing Excel-based simulation to visualize the sensor's output and evaluate its effectiveness. Our findings demonstrate the 3T APS architecture's potential to deliver superior image quality, highlighting its suitability for diverse imaging applications. The analysis of the 3T active pixel circuit simulation results reveal critical insights into the circuit's phototransduction efficacy and temporal response characteristics. The simulation output of the 3x3 pixel array circuit further delineates the performance of the APS with precision, indicating a high charge transfer efficiency and effective reset mechanism. The Excel visualization provides a binary image with stark contrast, reinforcing the sensor's potential in applications requiring binary decision-making. This study contributes significantly to the field of CMOS image sensor technology, offering a scientific and practical perspective on the development of next-generation sensors for applications in consumer electronics, automotive, and medical imaging. The research provides valuable insights into the sensor's operational dynamics and performance characteristics, paving the way for future advancements in sensor technology.

The source of power instability is one of the numerous variables contributing to the low efficiency of single-phase solar inverters. Power instability directly impacts the energy conversion efficiency of the photovoltaic system. The key to solving this issue is power decoupling technology, which effectively improves the working efficiency of photovoltaic inverters by separating the mutual interference of reactive power and active power and balancing AC and DC power conversion. This article, based on existing literature and research results, focuses on analyzing the two main problems of photovoltaic inverter power instability: the power coupling problem, where reactive power and active power interfere with each other, and the unbalanced AC-DC power conversion causing secondary voltage ripple problems on the DC side. By reviewing the causes, impacts, and power decoupling solutions of these two problems, this article further compares the advantages and disadvantages of different power decoupling methods. Additionally, the article discusses the future development trend of power decoupling technology, and the challenges faced when studying new power decoupling strategies, aiming to provide valuable reference for further research and optimization of photovoltaic inverters.

Tensor networks can describe entity objects from multiple dimensions to more accurately reflect all the features of the entity. There are a large number of physical devices in the power grid, with multiple basic attributes and frequent updates of device status. The use of tensor network models can effectively characterize the equipment in the power grid and the operational situation of the power grid. This paper proposes the use of tensor network models to analyze the equipment and dynamic characteristics of the power grid. Construct a tensor network model from multiple dimensions such as device type, voltage level, affiliation, and device status. In response to the large number and scale of equipment in the power grid, analyzing the characteristics of concentrated application scenarios, reducing the dimensions of tensor network models and highlighting commonly used data description dimensions to meet the needs of power grid applications. Based on this model, carry out application verification such as power supply reliability analysis and energy loss analysis. Extract corresponding data resources from different coordinate dimensions, and then calculate the inner product of the data from different dimensions. Based on the calculation results, perform neural network correlation analysis on different data resources, and solve the analysis results and solutions.

Wind power generation has emerged as a pivotal player in the global transition towards sustainable energy sources, garnering attention due to its renewable nature and widespread availability. However, the inherent variability of wind speed and other unstable factors in wind energy generation give rise to fluctuations in voltage and voltage flicker, significantly impacting the stability and quality of the electricity grid. This comprehensive analysis delves into the origins of voltage fluctuations and flicker induced by wind power generation, elucidating their detrimental effects on grid reliability. Moreover, it scrutinizes existing methods for detecting and mitigating these voltage issues, evaluating their efficacy and suitability across diverse environmental contexts. As wind power continues to evolve, this study also offers insights into future trends and technological advancements in wind power generation. By addressing the challenges posed by voltage fluctuations and flicker, this research contributes to the optimization of wind energy integration into the grid, fostering sustainable development in the renewable energy sector.

In order to improve the intelligent analysis effect of structured data of on-line monitoring of high voltage cable, this paper analyzes the monitoring process of high voltage cable with artificial intelligence technology, constructs a structured module with intelligent data analysis function, and analyzes the development of RPA process component, develops the overall communication framework with low power MQTT communication protocol and NB-IOT module. In order to improve the efficiency of data transmission and reduce the power consumption in the process of data transmission, this paper implements data rms calculation, invalid data elimination and data delay transmission design. Through the experimental analysis, it can be seen that the monitoring device can meet all the requirements of cable operation monitoring on the basis of low power consumption design, and can achieve long-term stable operation under the power supply mode of solar panel and battery.

In addition to reducing energy consumption and mitigating negative environmental impacts, wind power systems are critical for enhancing energy supply efficiency and reliability. However, the harmonic issues introduced by these systems present challenges to the stability and operational efficiency of power systems. Analyzing the interaction characteristics of harmonics and the power grid, establishing a reasonable harmonic source model, and developing new harmonic suppression technology have garnered extensive attention in the industry. This paper introduces the main types of wind turbines, among which the doubly fed fan is the most widely used. Subsequently, it briefly describes the mechanism of harmonic generation in wind power generation systems and explains the adverse effects of harmonics on power systems. The paper then summarizes the current common harmonic source models, including time domain, frequency domain, and data fitting models, and analyzes and compares these models, outlining their respective advantages, disadvantages, and application scopes. Furthermore, the paper reviews the current harmonic suppression technology and research progress in wind power generation systems, encompassing passive filters, active filters, static synchronous compensators, and other solutions. Finally, based on existing harmonic analysis and suppression technologies, the paper identifies the current challenges in harmonic research and application in wind power generation systems and proposes future research directions for harmonic control.

Weather temperature prediction is widely used in the fields of environmental management and disaster warning. Temperature prediction based on meteorology is more expensive and has a shorter time horizon. This paper proposes a weather temperature prediction method based on Long Short-Term Memory (LSTM) and Transformer to predict future temperature based on historical temperature information. First of all, Seasonal and Trend decomposition using Loess (STL) is used for feature decomposition to segment weather trends and noise. Then, feature extraction and prediction are carried out based on the LSTM-Transformer hybrid model. When the model is trained on real weather data, the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) reach 0.1902 and 0.2558 respectively, which is significantly better than the independent LSTM and Transformer models. These results show that the integration of LSTM and Transformer architecture can effectively improve the forecasting capability of meteorological time series analysis. Finally, the future research direction is prospected.

The demand for environmental protection has led to a continuous upsurge in the global capacity of wind power. However, in the process of gradually replacing traditional synchronous generators, the low-frequency oscillation problem has gradually come to the forefront and become an important factor affecting the stability of the power system, attracting the attention of many researchers. Especially in the field of Doubly Fed Induction Generators, research on the suppression of low-frequency oscillation has become a hot-spot in recent years. In this paper, several typical suppression strategies based on Power System Stabilizers are introduced. Additionally, this paper discusses recent adaptive suppression strategies for wind turbines and methods of designing new controllers. Furthermore, various low-frequency oscillation suppression methods and their effects are compared and analyzed. Finally, the paper summarizes and anticipates the development of low-frequency oscillation suppression.

The efficiency of regenerative braking energy recovery in electric vehicles is closely related to their operational performance, a BLSTM model was established to calculate the differences in energy consumption of electric vehicles under different driving styles, and the specific energy consumption of electric vehicles was judged according to the differences in energy consumption. The IIDP algorithm calculates the recovery efficiency of regenerative braking energy in electric vehicles and selects the most suitable recovery strategy. Combined with the IIDP-BLSTM model, it analyzes regenerative braking energy recovery strategies based on driving styles. Error test and simulation comparison experiment proved that the IIDP-BLSTM model studied in this paper had stronger energy saving in regenerative braking energy recovery of electric vehicles. And has good optimization performance.

In the current era, the automotive sector stands as a cornerstone of the national economic framework. The progress and breakthroughs in automotive technology have ushered in an era of comfort and affluence. However, the prevalent use of automobiles also yields substantial emissions of deleterious gases, contaminating our living spaces and depleting the planet's finite and precious oil reserves. Addressing the energy and environmental challenges posed by the automotive industry, while simultaneously fostering its growth, presents a critical scientific and technological conundrum that demands resolution within the automotive domain and indeed across society at large. Consequently, the pursuit of new energy vehicles has become an imperative for automakers. Amid the trio of prevailing technological pathways for new energy vehicles, hybrid technology emerges as the most developed, boasting superior overall performance, and has thus garnered the favor of the automotive industry. The control algorithms and real-time simulation of hybrid electric vehicles have consistently been at the epicenter of automotive industry research. However, the most widely used control strategy is still based on expert experience. Because the hybrid electric vehicle has a complex structure, and its operating conditions are usually unknown, it is not easy to design an adaptive and self-renewal control algorithm applied to the hybrid electric vehicle. Based on the deep reinforcement learning algorithm, this paper studies the control algorithm of parallel hybrid electric vehicle and real-time simulation test. The comparative experiments were carried out under NEDC working condition. The experimental results show that the control algorithm based on deep reinforcement learning has better energy-saving and control effect than the other two algorithms under the condition of full learning and training, and the algorithm has the potential to realize the general algorithm because of its self-adaptive and self-learning characteristics.

This paper mainly designs traffic lights for intersections and crosswalk. In order to ensure traffic safety and order and reduce the occurrence of traffic accidents, it is necessary to design a traffic light control system based on PLC control at traffic intersections. Because the PLC control system is easy to modify, easy to operate and easy to use, this paper comprehensively uses Mitsubishi PLC timer instruction, counter instruction, comparison instruction, touch screen, etc., to realize the control program design of traffic signal lights. The system is not only simple and practical, cost-effective and easy to repair, but also can automatically control of traffic lights, and the results show that the program can stably and reliably achieve dual mode control of intersection traffic lights and crosswalk traffic lights.

The risk of icing on power lines in areas with high rainfall is increasing due to extremely cold weather, which will pose a huge challenge to power supply work. The causes of icing on ultra-high voltage power lines are closely related to weather, terrain, temperature, etc. The high randomness of its occurrence makes it difficult to predict. A prediction method for icing disasters on ultra-high voltage transmission lines based on extemporization feature analysis is proposed in this paper, which analyzes the faults of ultra-high voltage transmission lines from the physical characteristics of power conductors and the changes in the external environment. Considering the mutual influence between the factors that cause icing, key factor extraction is used to sort meteorological, temperature, humidity, geographic and other data, and weight values are assigned to various indicator data based on the sorting results. Construct a multi factor comprehensive impact sequence data prediction model to incorporate the above indicators into the input parameters of sequence prediction. Based on the training of historical data using support vector machine algorithm, the fitting method of the ice cover risk prediction model mentioned above is improved, and the best fitting coefficient is found to improve the prediction accuracy of the model. Due to the comprehensive consideration of various factors such as weather, high precision terrain, and environment, this model predicts the occurrence patterns of icing disasters from both time and space dimensions, and the prediction results of icing disasters are more closely related to the actual situation.

Electrical equipment is an indispensable part of the production process, and real-time monitoring of it is crucial to prevent equipment failures, improve production efficiency, and ensure production safety. In response to the current problems in electrical equipment status monitoring, this article adopts a real-time monitoring system for electrical equipment status based on Ant Colony Optimization (ACO) algorithm. The system fully explores the operational status and expert experience of electrical equipment during operation, and designs heuristic information to guide ants in finding more valuable solution space. Improvements can be made to the ACO algorithm to better identify potential anomalies. Research heuristic information guidance methods that integrate the characteristics of electrical equipment, fully tap into the operational status and expert experience of electrical equipment during operation, and design heuristic information to guide ants in finding more valuable solution space. On the test dataset, the recall rate of the original ACO algorithm is between 73% -76%, and the maximum recall rate of the improved ACO algorithm is 84%. The method described in this article helps to improve recall rates and fully realize real-time monitoring of electrical equipment status.

Ultra high voltage power lines transmit large-scale electricity to economically developed areas and are currently an important carrier for long-distance power transmission. High voltage overhead lines are one of the main modes of power transmission. Due to the influence of natural environment and human activities, the transmission channel space faces various risks, and it is necessary to promptly investigate and alert potential risks. In response to this issue, this paper proposes a method for identifying anomalies in power grid transmission channels based on spatial data. Real time monitoring data of transmission channels is obtained through acquisition devices such as cameras and locators. Then, the image, video, geographic information and other data are associated and mapped, and a data model of the transmission channel is constructed with the device as the description core. Under extreme weather conditions such as strong wind, rainstorm and snowstorm, the power transmission channel is disturbed by floating objects and light objects in the nature, causing leakage, short circuit and other problems. Based on the established data model, simulate the movement trajectory of various floating objects that may occur, and then simulate and calculate their various impacts on the safety of transmission channels. Digitally deduce the scope of impact and the secondary accidents caused and determine the impact range of transmission channel risks caused by this event on power supply capacity failures.

This article presents an in-depth examination of the architectural and optimization strategies employed in the development of a four-bit absolute value comparator, emphasizing its pivotal role in elevating the energy efficiency and operational performance of digital processing apparatuses. In an innovative approach, the study melds the principles of static Complementary Metal-Oxide-Semiconductor (CMOS) with those of pass-transistor logic circuits to devise comparators that furnish both single-bit and triplet-bit outputs. This amalgamation leverages the intrinsic benefits of both circuit types, aiming to optimize energy consumption and enhance performance metrics. Through meticulous analysis and juxtaposition of two distinct comparator models, the research underscores the criticality of adept circuit sizing and the meticulous adjustment of supply voltages. These factors are paramount in diminishing energy requisites without compromising the integrity of computational accuracy or the throughput of the systems. The findings elucidate that strategic optimizations can lead to substantial reductions in power consumption, all the while either preserving or amplifying the system’s performance capabilities. Such enhancements are pivotal for the proliferation of electronic devices that prioritize energy conservation alongside high functionality.

The advancement of neural prosthetics and neuroscience research has promoted the development of spike sorting. As an important part of spike sorting algorithms, people are interested in designing and optimizing a high-performance absolute-value detector with low delay and low energy consumption to adapt to and meet the requirements of hardware with accuracy and protection of related tissue. The paper aims to design a four-bit Absolute-Value Detector by using CMOS circuit and calculate the minimum delay of critical path and total energy consumption by using the Logic Effort Theory. In this way, a 51.8 FO4(1V) and 70.1 Eu(1V) detector could be achieved. Further discussion is needed to explore the individual effects of sizing and VDD scaling on delay and energy consumption. Ultimately, calculations should be conducted to determine the energy consumption at 1.5 times the minimum delay when both sizing and VDD scaling are applied to seek optimization. A 77.7 FO4 and 37.7 Eu Absolute-Value Detector is achieved finally. In conclusion, by designing a four-bit Absolute-Value Detector, this paper provides the fundamental ideas and methods for designing and optimizing multi-bit absolute-value detector.

In non-isolated grid-connected photovoltaic systems, system losses are increased by common mode leakage currents. Common-mode leakage current generates current harmonics, causes electromagnetic interferences and threatens personal safety. Therefore, common mode leakage current suppression is an important technical issue. First of all, this paper analyzes the causes of leakage current and then explains the most common leakage current suppression method by the mechanism of typical inverter topology, i.e., constructing a new continuity loop to disconnect the photovoltaic cell side from the AC grid side during the continuity stage, and combining it with the switching modulation method to keep the common-mode voltage unchanged, thereby effectively inhibiting the source of leakage current. Subsequently, the new topologies proposed by domestic and foreign scholars to suppress leakage current, namely the three-phase H7 voltage source topology and NDPGCI-6 topology, are reviewed, and their working principles are analyzed and compared regarding the effectiveness of leakage current suppression. Finally, these new topologies are summarized, and the feasibility of their application in practice is examined.

With the rapid development of information technology and social economy, people pay more and more attention to the function of robots in data center. This paper analyzes the problems existing in manual inspection. proposes four kinds of robot inspection schemes, presents the comprehensive architecture design of data center inspection robots, focuses on the path planning strategy based on multi-sensor fusion, and completely presents the architecture and functions of intelligent inspection management platform. The research content of this paper is of great significance to the future development of data center.

This article is the design of Python face recognition based on Google Mediapipe framework and STM32F103C8T6; data transmission and calculation of STM32F103C8T6 Scontroller and vehicle camera. Through this experimental data prove that this design realizes the data transmission calculation from the on-board camera and the on-board computer and STM32F103C8T6 microcontroller, and generally realizes the design task of controlling NG995 steering gear through STM32F103C8T6 microcontroller.

Smart city monitoring system is a comprehensive, efficient and real-time intelligent monitoring system for monitoring urban traffic, environment and safety. It mainly uses the Internet of Things and artificial intelligence technology to realize the real-time monitoring and analysis of urban traffic, environment, safety and other aspects, and provides intelligent early warning and response solutions. It integrates a variety of perception technologies, data analysis and information communication technologies to realize real-time monitoring and intelligent management of all aspects of the city, and improve the efficiency, security and sustainability of the city operation. Through the intuitive visual interface, we can easily understand the operation status of the city, as well as the trends and changes of various data indicators, provide intelligent decision support, and help to develop more scientific and effective city management strategies. Thus, a comprehensive urban monitoring network is established to collect and analyze the data of all aspects of the city, including but not limited to transportation, environment, energy, etc, to prevent and respond to security incidents, so as to improve the overall safety level of the city. Based on big data analysis and artificial intelligence technology, it provides intelligent decision support system to provide scientific and accurate decision-making basis for city managers. By providing more efficient urban services to improve residents' quality of life and experience, such as intelligent transportation and intelligent waste management. By optimizing resource utilization, reducing energy consumption and environmental protection, we will promote sustainable urban development. The construction and application of smart city monitoring system will promote the development of related technologies and industries and promote the innovation and growth of urban economy.

With the rapid development of remote sensing technology, the integrated platform of micro synthetic aperture radar (Mini Synthetic Aperture Radar, Mini SAR) and multi-rotor unmanned aerial vehicle (Unmanned Aerial Vehicle, UAV) has shown great potential in disaster monitoring, environmental assessment and other fields. This paper focuses on the design, hardware integration, software function development and test experiment of the integrated micro-synthetic aperture radar and multi-rotor UAV remote sensing platform. The framework design of the integrated remote sensing platform is proposed, the selection and integration method of hardware components are introduced in detail, and the design concept and implementation process of software function are emphasized. Finally, the stability and practicability of the system are verified through the field test experiment, which provides the theoretical basis and practical guidance for the subsequent research and application.

As an excellent insulating medium, SF6 gas has many advantages and has been widely used in electrical equipment. However, with the introduction of relevant regulations on carbon emission reduction and greenhouse gas emission reduction such as the "dual carbon" target, the pace of research on the replacement of SF6 gas insulation for high-voltage power equipment has been accelerated. At present, clean air, as one of the environmentally friendly alternative gases, has gradually begun to be used in electrical equipment. Therefore, the study of clean air insulated suspension discharge is of great significance for the operation and maintenance of electrical equipment. In this paper, a photoelectric joint detection platform was first built to detect the pulse current, ultraviolet band, visible light and infrared band signals of suspension discharge in clean air insulation. The development process of suspension discharge in clean air insulation was studied, and the change law of each signal was analyzed respectively, providing a reliable basis for fault diagnosis and pattern recognition of clean air insulation electrical equipment.

The design drawings in power grid engineering construction usually need to be transferred to the equipment operation and maintenance stage. However, due to the differences between the actual construction process and the design drawings, manual adjustment has the problems of large workload and low efficiency. This article proposes the intelligent registration and reconstruction of power grid engineering data twins, which solves the problem of the lack of grid-based 3D point cloud data for modifying and adjusting the design drawings of substations in related technologies and improves the accuracy of operation and maintenance control after transferring the 3D design drawings to the operation and maintenance stage. This article studies the application of transfer model data in unmanned aerial vehicle inspection business for power transmission and transformation, improving the safety of unmanned aerial vehicle autonomous inspection and enhancing the reusability of transfer models by operation and inspection departments.

With the development of Industry 4.0, the operation training of industrial equipment is more and more combined with information technology, and the application of mixed reality technology in industrial training has become a trend. Based on HoloLens equipment, this paper designs and develops a pumped-storage power station equipment training system based on mixed reality with MRTK and Unity 3D as the development engine. The training system we developed is deployed on HoloLens devices and tested and evaluated. As a combination of virtual reality and augmented reality, MR technology provides new ideas and possibilities for the research and application of equipment training system in pumped storage power station. The test results show that the system runs smoothly and achieves a good effect of virtual-real fusion, users can use diversified interactive means to operate the training system naturally, the realistic experience of device operation is stronger, and the training effect is remarkable.

With the transformation of global energy structure, pumped storage power station, as an important energy storage technology, plays a key role in the power system. However, the efficient operation of power plants depends on regular overhaul and maintenance, which faces many challenges. The application points of smart wearable devices in the maintenance of pumped storage power stations mainly include equipment selection and customization, system design and integration, application scenario analysis, etc. The results show that MR technology can significantly reduce maintenance time, reduce accident rate, shorten training cycle, and reduce annual maintenance costs. In addition, the maintenance work supported by MR technology has also received higher satisfaction from maintenance personnel. This study provides an empirical analysis for the application of MR technology in the maintenance of pumped storage power station and puts forward some suggestions for the future development direction.

This paper introduces a novel AI-based automated circuit design optimization method aimed at enhancing design quality and efficiency. The proposed modular architecture integrates artificial intelligence with circuit design expertise, comprising key components for data processing, optimization algorithms, circuit simulation, and feedback mechanisms. We implement and test this system on a high-performance computing cluster, utilizing a comprehensive test suite of both analog and digital circuits. Our method demonstrates significant improvements in critical performance metrics, including power consumption, delay, and area, with enhancements ranging from 10% to 30% compared to traditional design approaches. The system's ability to generate multiple optimized design candidates under various constraints provides designers with flexible trade-off options. Furthermore, the high degree of automation in our approach substantially reduces manual effort and accelerates the design process. These results underscore the potential of AI-driven techniques to revolutionize integrated circuit design methodologies.

The escalating demands of 5th-generation mobile communication technologies (5G) for enhanced network capacity, energy efficiency, and quality of service drive the need for innovative network management and architecture designs. The Downlink-Uplink Decoupled access network (DUDe) represents a transformative approach, enabling independent optimization of connections to different Base Stations (BS), significantly boosting flexibility and efficiency in urban environments characterized by high user density and fluctuating traffic demands. This paper introduces a comprehensive optimization framework that employs Successive Convex Approximation (SCA) to refine power allocation, user access, and throughput in 5G networks. Through extensive simulations, the results demonstrate that this method notably surpasses traditional approaches by enhancing throughput and user satisfaction, marking a significant step toward next- generation network capabilities.

With the growth of global energy demand and the improvement of environmental protection awareness, the efficiency and safety of the maintenance of pumped storage power station as a key component of the power system is particularly important. This paper deeply discusses the application of mixed reality technology in the maintenance task of pumped storage power station, aiming at improving the modernization and intelligence level of maintenance work. By analyzing the existing maintenance process and combining with the characteristics of MR technology, this paper proposes a set of MR-based maintenance process optimization scheme, including remote fault diagnosis, remote maintenance assistance, auxiliary maintenance training and maintenance safety education. The results show that the application of MR technology significantly improves the maintenance efficiency, reduces the cost, and enhances the training effect, which has an important impact on the safe and stable operation of the power plant and the reliability and economy of the power system.

With the development of science and technology, Unmanned Aerial Vehicle (UAV) technology has played a good role in modern logistics, communications, military, agriculture, disaster relief, and other fields. However, the lack of endurance capacity of drones restricts their development in various fields. Therefore, research on drone charging technology has become a key factor in facilitating continuous breakthroughs in drone technology. To improve the accuracy of automatic charging path planning for drones operating in multi-drone states in substations, this paper proposes a neural model for multi-drone full-area reconnaissance path planning based on a neural network coupled with a deep reinforcement learning algorithm. To test the accuracy of the flight path for automatic drone charging in multi-drone areas obtained through neural networks, greedy strategies, and experience replay designed in this paper, simulation experiments were conducted. The results showed that the automatic drone charging model for substations designed based on the neural network and deep reinforcement learning algorithm reduced the charging path length by 8.9% compared to traditional drone path planning algorithms. It lowered energy consumption during the drone's charging flight path, and the algorithm converged faster, demonstrating better path planning performance.

With the continuous growth of power demand and the increasingly complex operating environment of power grid, the traditional power grid management model has been difficult to meet the needs of modern management. Therefore, by integrating big data technology and AI (artificial intelligence) algorithm, this study aims to build a high-performance and intelligent power grid management platform, namely, DIPGVRSP (Digital Intelligent Power Grid Virtual-Real Synchronous Platform). During the construction of DIPGVRSP, a layered architecture design is adopted, including data acquisition layer, data processing layer, data analysis layer and application layer. In order to further optimize the performance and efficiency of DIPGVRSP, this study puts forward a number of strategies, including data processing efficiency optimization, intelligent scheduling and optimization, and security and reliability improvement. By introducing parallel processing and distributed computing technology, the data processing speed is significantly improved. At the same time, based on historical data and machine learning model, a predictive scheduling algorithm is developed to realize intelligent scheduling and optimization of power grid. The experimental verification shows that DIPGVRSP based on LSTM has achieved high accuracy in power grid load forecasting, and the forecasting performance is improved by about 20% compared with the traditional linear regression model. This result not only verifies the effectiveness of DIPGVRSP, but also provides strong support for the further popularization and application of digital intelligent power grid technology.

The three-in-one electric drive system assembly of new energy vehicles plays a key technical role in the development planning of China's new energy vehicle industry, so more stringent requirements are put forward for the noise, vibration and harshness (NVH) level of the three-in-one electric drive assembly for vehicles. The noise source and transmission path of the three-in-one electric drive assembly are analyzed, and the relation between the modal frequency and stiffness of the motor controller cover plate and the noise of the electric drive assembly is studied. The modal frequencies of the upper controller and the controller cover plate in the electric drive assembly and the relation between them and noise are investigated. The modal optimization solution model of the electronic control cover plate is established, and the Optistruct module of Hyperwork software is used to obtain the best topology. This includes increasing the thickness of the electronic control cover plate, increasing the circumferential solar rib to increase the modal frequency and stiffness, and also increasing the acoustic wrap to optimize the vibration and noise of the electric drive assembly. After the stiffening treatment, the free mode frequency of the first-order assembly is increased by 209Hz, the increase is 22.9%, and the amplitude is significantly reduced. According to the relationship between stiffness and natural frequency, it can be seen that the stiffness of the optimized rear cover plate is significantly increased. Finally, we carry out experiments to verify the remarkable optimization effect of this scheme on the vibration and noise of the electric drive assembly. Through the comparison of the order slices of typical order noise, we observed that the 22th order of the first gear of the reducer, the 5.94th order of the second gear, and the 24th and 48th order of the motor noise, the amplitude of the corresponding vibration acceleration of the cover plate was reduced by 6 ~ 17dB, and the noise level of the electric drive assembly was effectively reduced.

The digital transformation of road and bridge maintenance management, as well as the research and application of intelligent detection technology and equipment, are key research directions for promoting the development of digital transportation construction. The good mobility, safety, and economy of Unmanned Aerial Vehicle (UAV) make it has good application value in bridge inspection, which can effectively make-up for the drawbacks of low efficiency, high cost, great danger, and blind area of traditional inspection methods. This paper summarizes and analyzes the advantages of UAV bridge inspection, and introduces key issues such as route planning, image acquisition, 3D reconstruction, defect identification and localization and visualization, and inspection management. Design and develop a bridge inspection management system based on drones, build a bridge intelligent inspection solution with full coverage, high efficiency, high precision and visualization, and provide a reference for the innovative application of unmanned intelligent equipment in highway inspection and maintenance.

To address the issue that traditional bearing fault diagnosis relies heavily on expert system, and its shallow machine learning method is not effective for bearing fault identification. This paper introduces a model for bearing fault diagnosis utilizing STFT-SWT and two-stream CNN-KNN.To begin with, the vibration signals from non-stationary bearings are transformed into high-frequency time-frequency representations, creating a two-dimensional depiction. These representations are then utilized by a Convolutional Neural Network (CNN) to achieve accurate bearing fault diagnosis. Finally, the fault diagnosis classification results are output by KNN. In order to solve the problem of uneven time-frequency resolution of short-time Fourier transform, it is combined with synchronous compressed wavelet transform. Experimental results using the bearing dataset from Case Western Reserve University indicate that the diagnostic performance of this method surpasses that of other traditional models and effectively enhances the diagnostic accuracy of bearings.

In order to improve the rolling bearing fault diagnosis accuracy, a rolling bearing fault diagnosis method with Variational Modal Decomposition (VMD), t-distributed domain embedding (t-SNE) and genetic algorithm optimized least squares support vector machine (GA-LSSVM) is proposed. Firstly, the bearing signal is decomposed into several intrinsic modal components (IMFs) by applying VMD; secondly, the time-frequency eigen indices of each IMF are calculated to compose the high-dimensional fault features, and the t-SNE algorithm is utilized for the secondary feature extraction, which reduces the influence of redundant features on fault diagnosis. Finally, the low-dimensional sensitive features are used as inputs to the GA-LSSVM model. Experimental analysis shows that the model based on t-SNE-GA-LSSVM has high fault diagnosis accuracy, good stability

With the wide use of Brushless DC Motor (BLDCM) in industry, traffic and home applications, higher requirements are put forward for the performance and stability of its control system. Although the traditional PID control method is simple and easy to use, it may not perform well in the face of nonlinear and time-varying BLDCM systems, especially in the face of external disturbances and parameter uncertainties. In order to improve the control effect of BLDMC system, this paper devotes itself to designing an advanced speed control system, which adopts sliding mode control as the main control strategy. SMC can effectively suppress uncertainties and external disturbances by introducing sliding mode surface, thus improving the robustness and dynamic performance of the system. However, the traditional sliding mode control will bring bad jitter phenomenon because of high frequency switching. In order to reduce the occurrence of jitter, this topic adopts the integral sliding mode method for motor speed control to ensure the accuracy and reliability of brushless DC motor control.

In response to the problem of insufficient adaptive ability of a single intelligent fault diagnosis model, this paper proposes a new fault classification method based on the Golden Jackal Optimization (GJO) algorithm optimized Least Squares Support Vector Machine (LSSVM). This method first utilizes adaptive noise complete empirical mode decomposition (CEEMDAN) technology to decompose and process the collected vibration signals, thereby obtaining a series of intrinsic mode components (IMFs). Subsequently, effective components were selected from these IMFs based on screening criteria, and the energy and waveform coefficients of these effective components were extracted to jointly form the feature set for fault diagnosis. Next, the GJO algorithm is used to Perfect the selection of kernel parameters and penalty parameters in the least squares support vector machine LSSVM, thereby establishing a GJO-LSSVM fault diagnosis model. Finally, input the feature vector matrix into the GJO-LSSVM fault diagnosis model for fault diagnosis and analysis.

Fault prediction and diagnosis of induction motor is very important for the smooth operation of industrial production environment. The most common fault of asynchronous motor is the short circuit between rotor broken bar and turns. However, it is difficult to collect the fault current signal of motor in actual industry. So, establishing an accurate and stable current signal prediction model is of great significance for fault prediction and diagnosis. This paper presents the prediction of motor fault current signal based on LSTM (Long short-term memory neural network), which provides important reference for the subsequent motor fault diagnosis technology.

In order to solve the problem of fault diagnosis of elevator worm gear box bearing, the method based on continuous wavelet transform and long short-term memory neural network is adopted. The collected data is first filtered by an infinite impulse response filter, and then the scale maps of the current signal and the vibration signal are obtained by continuous wavelet transform and spliced at the same time, and finally they are put into the long short-term memory neural network for analysis and training. The experimental results indicate that the accuracy of the proposed approach for the fault diagnosis of elevator worm box bearings can reach 99.85%.

The PCB sliding coil motor without iron core has the advantage of no teeth interlocking, no positioning force, lower thrust density, no attraction between the stator and the rotor, which leads to a smoother thrust generation, reduced thrust fluctuation, and correspondingly lower vibration and noise. These are its significant advantages, but it also has the disadvantage of complex manufacturing process, complicated bone structure design and winding fixing, and difficulty in installation in applications with limited installation space. In view of this problem, this paper designs a PCB sliding coil motor with a thrust of 6.4N. The motor uses a PCB board to replace the epoxy resin used to fix the winding, which can reduce the size of the motor. The PCB winding has the advantages of good manufacturing process, high precision, low cost, and easier mass production. This paper conducts research on the electromagnetic design of the PCB sliding permanent magnet synchronous linear motor and carries out design work on the PCB sliding permanent magnet synchronous linear motor. Initially, the structure and operating principles of the linear motor are elucidated, followed by the design of the motor in accordance with specified requirements. Upon determining the motor's dimensions, a model is constructed and simulated using ANSYS Maxwell software. The simulation outcomes are then analyzed to validate the accuracy of the design.

Aiming at these difficulties in early warning of weak fault signals and capture of weak features of rolling bearings, a method for early warning and diagnosis of weak fault of rolling bearings was proposed. Firstly, envelope components were extracted from bearing vibration signals, and the Improved Signal Noise Ratio (ISNR) was acted as the optimization objective function. This delay feedback stochastic resonance optimal model (TFSR) was constructed and combined with the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and this intrinsic mode component (IMF) screening, the first reconstructed signal was obtained, and its ISNR at the fault characteristic frequency was calculated and compared with the preset initial threshold to determine whether an early warning was issued. Then, if an early warning is issued, the vibration signal is denoised by multi-wavelet adjacency coefficient adaptive threshold method, and the denoised signal is screened by fast spectral kurtosis and ensemble empirical mode decomposition method to obtain the second restored signal. Then the improved rapid spectral coherence method is used to obtain this corresponding enhanced envelope spectrum to realize fault diagnosis. Finally, XJTU-SY data set is taken as an example to confirm the feasibility of this approach.

This paper proposes an amalgamation of blockchain and Interplanetary File System (IPFS) incentive approach for distributed electric vehicle (EV) charging. The incorporated blockchain technology with IPFS protocol ensures cost- efficient and reliable data transactions between EVs and Charging Stations (CSs). We have integrated a fifth generation (5G) communication network to ensure low latency and highly reliable communication between EV and CS. Furthermore, the proposed blockchain and gamification approach employs a conditional zero sum (C-ZeSu) game theory to optimize the incentive for EV charging. The performance analysis of the proposed incentive approach is evaluated and implemented over the Remix Integrated Development Environment (R-IDE) by deploying the smart contracts of formulated C-ZeSu game in terms of various performance evaluation metrics such as convergence and transaction efficiency analysis, 5G New Radio (NR) bit error rates.

This paper introduces a new method for traffic anomaly identification within the Faster R-CNN framework by combining the DenseNet module with the Region Proposals (RPs) processing block. The suggested approach seeks to improve feature extraction and RP creation in aerial imagery or CCTV frames to get beyond the drawbacks of the current methods. Discriminative features are extracted through DenseNet, and the RPs block makes it easier to provide a wide range of accurate and varied area proposals. Contrast correction techniques are applied to overcome the difficulties presented by low-resolution and low-contrast images frequently seen in surveillance scenarios. Comparing experimental results to state-of-the-art techniques, one can observe notable resilience and detection accuracy increases. The suggested method works exceptionally well in various traffic scenarios and environmental circumstances, indicating its potential for practical implementation in traffic surveillance systems. Overall, the study contributes to advancing the efficacy and reliability of anomaly detection in urban traffic environments.

In order to improve the functional testing requirements of the new energy vehicle design process, the test system of automotive electronic control equipment, which is based on LabVIEW simulation platform, integrates the functional test of electronic control equipment and the network test of electronic control equipment, and provides a powerful software and hardware platform for the test of automotive electronic control system. The real-time control system proposed by this paper mainly completes the functions of data collection, model calculation, test piece and dynamometer motor control, which is the core part of the whole test system. After that, the characteristic curve of the motor is obtained by fitting the test data by numerical calculation. In addition, the off-line simulation of the motor model drive system is needed before the test system is used. The experimental results show that the data of acceleration and deceleration performance of the test platform proposed in this paper are basically consistent with the actual data. Therefore, the model proposed in this paper can meet the requirements of electronic control function test of new energy vehicles and can provide reference in the follow-up functional test of new energy vehicles.

Urbanization is accelerating, which has highlighted the need for plant conservation in a number of areas, including agricultural output, urban greening, and horticulture maintenance. However, there are a number of significant disadvantages to the traditional watering methods used in these locations, such as waste of water resources, low efficiency, and environmental degradation. As such, the creation and application of intelligent autonomous watering robots has become an important field of current research. In order to tackle the urgent problems related to plant watering, this paper provides a thorough analysis of the mechanical structural design of such robots. This research's main goal is to present a thorough explanation of the mechanical design of an intelligent, autonomous watering robot. An inventive, autonomous watering robot with a unique mechanical construction is painstakingly researched and built using SolidWorks modeling technology. Because of its special design, the robot can move with amazing freedom and agility through agricultural areas, green spaces, and gardening fields. It can also accurately irrigate plants by combining six-axis manipulator operation with orbital mobility. Its mechanical structure also demonstrates remarkable flexibility and adaptation, which enables it to meet the different watering needs of different plant species and environmental circumstances while remaining sturdy and dependable. The study illustrates the observable advantages of using these robots in real-world settings through extensive empirical testing. The robot notably lessens the labor-intensive manual workload related to watering plants, which boosts output and effectiveness. Moreover, using it results in a significant reduction in the waste of water resources, supporting environmental preservation initiatives and sustainable development objectives. To sum up, the innovative development of intelligent automatic watering robots signifies a noteworthy progress in the realm of plant preservation. These robots are vital to the advancement of sustainable development and environmental stewardship because they provide an efficient solution for agricultural production, urban greening, and horticultural maintenance.

This study conducted an in-depth analysis and optimization of the capacity configuration of CSP power plants by applying an improved intelligent optimization algorithm. An actual CSP power plant located in a high sunshine area was selected as a case, and optimization analysis was performed using solar radiation data, meteorological data, and power demand data. By comparing the performance of genetic algorithm (GA), particle swarm optimization algorithm (PSO), and ant colony optimization algorithm (ACO), it was found that the PSO algorithm performed best in terms of convergence speed and optimal fitness, and the GA and ACO algorithms also showed high stability in terms of final fitness values. The effectiveness and applicability of intelligent optimization algorithms are shown from the actual case analysis, and the optimized scheme of capacity configuration will obviously promote the efficiency of the power plant operation. The comparison of the annual change characteristics of the data and the optimization results shows that the operating efficiency of the CSP power plant can be significantly improved by reasonably selecting and improving the intelligent optimization algorithm, indicating the practical application value and potential broad application prospects of the intelligent optimization algorithm in the capacity configuration optimization of CSP power plants.

This research focuses on implementing machine learning in optimizing parameters and controlling power electronic systems and suggests a fusion optimization technique using neural network and genetic algorithm. Initially, a mathematical model is developed for optimizing parameters in a power electronic system, with the utilization of genetic algorithm for optimization. Next, a system for adaptive control is created, incorporating the use of neural network for both parameter estimation and optimization of control strategy. By analyzing a specific case of an electric vehicle charging system, it was found that this approach greatly enhances the system's charging efficiency and stability. The results from the experiment confirm that this method is effective and practical for optimizing power electronic systems, offering fresh ideas and techniques for system design and control improvement.

Typical PI control and passive control can not guarantee the reaction rate of the system when the system is perturbed because of the fixed parameters of the internal control structure, and can not maintain a good control effect. In order to deal with the problem of fixed data under passivity-based control, another control can be added based on the Euler- Lagrange model of the inverter system, the chattering of sliding mode variable-structure control (SMC) can be restrained by improving the anti-disturbance ability of the whole system and optimizing its reaching law, so the system can get better output characteristics. On the basis of MATLAB/Simulink platform, the inverter system simulation module is built to carry out simulation, by comparison and analysis, it is proved that under the consistent simulation conditions, the passive and sliding mode control has stronger anti-disturbance and better response speed.

Off-grid microgrids, characterized by significant dynamics and distinct constraints, pose challenges to scheduling practices. This paper advances a research approach to optimize the scheduling of off-grid microgrids based on simulated annealing algorithm(SA). Our study seeks to initially disorganize power source allocation across the microgrid to create a new solution for SA, then accordingly calculate and allocate loads using the greedy algorithm. The adoption of the new load allocation follows the principle of simulated annealing, eventually leading to the optimal solution. We implement scheduling experiments on test microgrids and compare our algorithm with conventional techniques such as linear programming and genetic algorithms. Comparison results demonstrate an effective decrease in total system costs with our algorithm, outperforming the classic methods on performance indicators such as system stability and cost efficiency. Lastly, system services with our algorithm prove to be close to those of state-of-the-art techniques, hence validating the potential advantages and applicability of the proposed method in optimizing off-grid microgrid scheduling.

Modern machinery often possesses complex structures and operates in harsh environments, making it prone to failures. Vibration signals, containing crucial information about equipment operation status, are frequently affected by significant noise interference during the collection process, which results in the crucial components being overshadowed by noise, making fault characteristics difficult to extract. To address these issues, a sparse decomposition algorithm based on successive orthogonal matching pursuit and morphological component analysis is proposed for extracting fault features from rolling bearing vibration signals. This method utilizes two threshold screening criteria based on statistical principles, exhibiting strong noise resistance and consequently better recovery capabilities for signals under strong noise interference. Experimental analysis confirms the effectiveness of this method in extracting fault features from rolling bearings.

With the rise of electric vehicles and electric bicycles and the growth of demand for new energy storage, the battery industry has gradually expanded to private enterprises, and the daily use of the battery has become wider. This design is divided into three parts: circuit design, component parameter calculation and simulink simulation. A filter circuit is added between the rectifier circuit and the buck chopper circuit to filter out a more stable DC output voltage, which is used as the input power of the buck chopper circuit and then supplied to the battery for charging. The simulation is carried out by Simulink, which verifies that the output voltage value and output current value reach the expected value.

In a multipath environment, the positioning receiver is unable to distinguish multipath signals, resulting in a decrease in the tracking accuracy of the receiver towards the positioning signal, which in turn affects the positioning performance. This paper proposes a pseudo-random code tracking loop algorithm for a positioning receiver based on the Multiple Signal Classification (MUSIC) algorithm, which improves the resolution of the positioning receiver to multipath. In the preprocessing stage, time-domain denoising and time-domain extension methods are adopted to make the relevant results in the receiver applicable to the MUSIC algorithm. Then, in the process of constructing the autocorrelation matrix, the frequency-domain truncated cyclic smoothing algorithm is adopted to further increase the signal-to-noise ratio and improve the efficiency of MUSIC pseudospectrum search. The test data shows that the above algorithm can distinguish the first path delay in multipath environments, improve the pseudospectrum peak by more than 50dB, and improve the tracking accuracy of the receiver on the positioning signal. The residual code phase in the same test environment is 38.9% higher than that of traditional receivers, thereby improving the positioning accuracy by 35.3%.

CK6150 precision CNC lathe can operate parts with high difficulty and high precision requirements and can also turn threads of various shapes and materials, so it is widely used. According to the technical specifications of CK6150 CNC lathe, select the appropriate FANUC 0i Mate TC CNC system, and reasonably select the four-position electric tool holder, and design the electric control system of the electric tool holder, the design is simple and reliable, and the protection measures are complete.

The rapid development of wireless local area network technology has led to the integration of wireless terminal devices into various aspects of our lives. An outdoor portable environmental monitoring system is developed. At present, most of the environmental monitoring instruments on the market are used in the detection of air, lake, ocean, river and other large-scale systems. Based on this, this article uses an embedded development platform and IoT technology to develop an Android interface in mobile terminals. Users can monitor the outdoor environment (such as temperature and humidity information, lighting information, PM2.5, etc.) in real-time through the network. The device is portable and low power consumption, high test accuracy.

This paper describes the use of commonly used digital chips to design and manufacture digital clock circuits. The clock circuit includes a pulse generation circuit, an hour, minute and second counting circuit, a display circuit, an hour, minute and second calibration circuit, and the entire chime circuit. Multisim software is used to simulate the entire circuit to determine the feasibility of the circuit. Finally, Litron EDA is used to realize the PCB. The appropriate PCB version is selected through Litron EDA, and the appropriate components are selected for physical installation, soldering and debugging, and finally the digital clock is realized. The PCB version is selected through Litron EDA, and the appropriate components are selected for physical installation, soldering and debugging, and finally the digital clock is realized. The appropriate PCB version is selected through Litron EDA, and the appropriate components are selected for physical installation, soldering and debugging, and finally the digital clock is realized. The appropriate PCB version is selected through Lichuan EDA, and the appropriate components are selected for physical installation, soldering and debugging, and finally the digital clock is realized.

In the rapidly advancing realm of information processing, the intersection of Field Programmable Gate Array (FPGA) technology and Convolutional Neural Networks (CNNs) has garnered considerable attention. This paper presents a comprehensive review of the evolution of FPGAs and CNNs, delving into the architectural intricacies of FPGAs and the compositional elements of CNNs, encompassing the convolutional layer, pooling layer, and fully connected layer. A particular emphasis is placed on analyzing the principles and functionality of the pipeline structure in FPGAs, elucidating its pivotal role in expediting the CNN inference process. The integration of pipeline architecture within FPGAs is a strategic design choice aimed at enhancing data processing efficiency. This structure allows for the execution of instructions across multiple stages concurrently, significantly boosting computational speed, a feature particularly advantageous for the extensive matrix operations characteristic of convolutional neural networks. Concurrently, CNNs have emerged as a powerhouse in deep learning, demonstrating remarkable success in image recognition and speech processing domains. These networks function by extracting features through sequential layer processing and executing classifications in the fully connected layer. When juxtaposed with traditional CPUs and GPUs, FPGAs reveal distinct benefits in CNN applications. The inherent programmability of FPGAs permits extensive customization, aligning seamlessly with diverse neural network architectures and achieving superior energy efficiency. However, it's crucial to note that CPUs and GPUs maintain certain advantages in general-purpose computing, particularly in handling complex logic and extensive data sets.

To address the issue of manual observation for aligning operations during the loading of the large target, a 3D depth camera was used to measure the position of the alignment target. The recognition of the alignment target is completed by selecting the fixed threshold algorithm and the contour-based minimum enclosing circle recognition algorithm in the position measurement process, and the measurement of the target position is realized by three-dimensional coordinate stereo solution. The experimental results indicate that the error range of position measurement after function fitting is within ±1mm. This meets the requirements for automatic alignment during the loading process of the large target.

In this study, we conducted in-depth research on the fusion of photogrammetric data based on the Unscented Kalman Filter (UKF) algorithm. By introducing the UKF algorithm into the process of electrical signal processing, we have successfully solved the problems of the traditional Kalman filter in terms of the lack of accuracy and stability. Principal component analysis and longitudinal and horizontal data fusion enhance the impact of photogrammetric data fusion. Through simulation experiments, the proposed method is verified to be more stable and accurate in terms of fusion bias and fusion result recall. This study provides useful guidance for the selection and optimization of data fusion algorithms in practical applications.

With the effective application of lasers in many aspects, fiber lasers attract increasing attention due to their excellent output results and material advantages. Among them, fiber lasers in specific wavelength bands have important application value in national defense, industrial processing, medical, and other related fields. However, there are currently few relative studies on the parameter design and relationship analysis of erbium-doped fiber lasers in specific wavelength bands. There is also a lack of model analysis on lasers. The issue of improving output power enters a bottleneck period. Based on the realistic physical model, this research report creates a simulation model of fiber laser. The fiber laser completes the simulation experiment of erbium-doped fiber laser in the 1500-1550nm band. It uses model simulation to calculate the output power and analyze the relationship between power and fiber length. The erbium-doped fiber laser in this research is designed for higher output performance. The results of this research create a generalized laser design model. The model can meet the design needs of solid fiber lasers in specific wavebands using various materials in various fields. The model can reduce the cost of experimental research. The results can also provide a high-performance design idea for the erbium-doped fiber laser within the 1500-1550nm band.

This paper presents a comprehensive model for detecting SIP-based man-in-the-middle (MITM) attacks, crucial for enhancing the security of Voice over Internet Protocol (VoIP) systems. Leveraging the Hidden Markov Model (HMM), the study introduces a dynamic behavior analysis framework that identifies deviations from typical SIP communication patterns in real-time, thereby improving the detection accuracy and efficiency. Additionally, the model incorporates robust encryption techniques and dynamic key management strategies to safeguard against both conventional and quantum computing threats. Through real-time monitoring and alerting modules, the system promptly responds to potential security threats, ensuring the integrity and confidentiality of the communications. By integrating enhanced encryption methods, such as Quantum Key Distribution (QKD) and Diffie-Hellman key exchange, the model provides a fortified defense mechanism that is particularly adept at mitigating the risks associated with MITM attacks. The proposed solution not only secures communication channels but also preserves the privacy and integrity of user data, offering significant security benefits to VoIP service providers and users alike.

As power systems become more intricate, traditional fault detection methods have difficulty keeping pace, especially in terms of accuracy and real-time performance. The recent advancements in deep learning technology have opened up new possibilities for solving these problems. This paper, leveraging deep learning technology, examines methods for detecting and intelligently predicting abnormal faults in power systems. First, we introduce in detail the basic knowledge of deep learning and its application status in power system fault detection. Next, by building and optimizing deep learning models, this study demonstrates how to effectively identify and predict abnormal faults in power systems. Specifically, we utilize a Convolutional Neural Network (CNN) to construct a model, experimentally verifying its efficiency and accuracy under various fault scenarios. In addition, this study also explores the application potential of deep learning technology in the fields of real-time monitoring of power systems, predictive maintenance, and energy management optimization. Finally, the paper summarizes the findings and discusses existing limitations and possible directions for future research.

To achieve high-speed performance of Fiber Channel over Ethernet (FCoE) transmission interface, a Direct Memory Access (DMA) controller based on FPGA is designed. The controller supports AXI4 bus protocol and takes Round- Robin strategy for priority arbitration of multi-channel selection (32 channels with eight priorities at maximum). In order to promote the transmission rates of discrete data, the controller supports Scatter/Gather mode based on list descriptors that are stored and analyzed by DMA controller other than the MPU. Therefore, the MPU’s efficiency can be promoted. Using ISim tool of Xilinx ISE development kit, the reading and writing function of the DMA controller is verified through BRAM and FIFO access on Xilinx-Virtex6 FPGA chip.

With the improvement of quality management system requirements year by year, the task side's requirements for product quality are becoming more and more stringent. How to improve and efficiently carry out quality work is the focus of our research. Through the implementation of three stages of quality control, that is, before, during and after the three stages of the whole system process quality control, reduce the amount of product rework, improve product quality. In this paper, the rework of printed circuit boards in the past year is taken as an example, data analysis is carried out, and the main reasons for repeated rework problems are summarized, which is used as the quantitative basis for pre-control of printed circuit boards production, and corresponding quality control measures are taken to solve existing quality problems, so as to ensure the rigor and effectiveness of quality control work.

With the increasingly high requirements of electronic products on the reliability of electrical equipment and the gradual development and wide application of crimping connection technology, this paper systematically introduces the crimping method, technical requirements and inspection standards of cord end terminals in cable assembly crimping from the aspects of the comparison of crimping and welding technology and crimping process of cable assembly. At the same time, an example of common terminal crimping is given to provide engineering reference for cord end terminal crimping process in subsequent cable assembly crimping, in order to improve product reliability, reduce production costs, and further improve product quality.

Aiming at the electromagnetic compatibility problems caused by the wiring of the electronic equipment of control system and the low efficiency of the traditional wiring design and wiring construction, a method of the wiring design of the electronic equipment of control system is proposed based on the introduction of the wiring design technology. Through the wiring design of electronic equipment of control system, the method to solve the electromagnetic compatibility problem of electronic equipment is given, and the wiring design efficiency and construction efficiency of electronic equipment are improved.

This paper fire safety is an important part of social public safety, although China has made a certain development in the intelligent fire detection technology, but there are still problems such as detection accuracy and response speed, environmental adaptability and cost maintenance. Based on the above problems, in order to improve the fire monitoring rate and reduce fire events, a multi-mode comprehensive fire detection and alarm system is designed. The design is designed by Keil 5, the principle is that the flame sensor and carbon monoxide sensor work at the same time, can detect the gas and flame at the same time, and combined with the data fusion, greatly improve the accuracy rate, reduce the false positive rate, and combined with the GSM module to solve the problem of unable to real-time feedback. At the same time, it can also control the pump, fan and window through the GSM module to effectively mitigate the fire situation.

With the continuous progress of society and the rapid development of technology, multimedia video processing technology has been widely applied in various fields. This paper presents the design of a functional video processing tool based on the Pango50K development board, capable of performing various image processing functions, including four-way stitching, arbitrary scaling, brightness and chroma adjustment, edge detection, and image translation in all directions. It aims to meet the diverse demands for multifunctional video processing. The system adopts a pipeline approach to image processing, ensuring the transmission speed of video streams and display frame rates.

With the increasing data scale and the increasing complexity of data processing requirements in the power industry, the demand for high-performance and low-power processor solutions is increasingly urgent. According to the characteristics of power business applications, this paper proposes a software and hardware co-design strategy of RISC-V (Reduced Instruction Set Computing-V) chip for power business applications. The strategy focuses on instruction set optimization, memory system optimization, parallel computing optimization and hardware accelerator design, aiming at improving the performance and energy efficiency of the chip and meeting the demand of large-scale data processing in the power field. Through case study and verification, the research results show the remarkable effects of this strategy in improving processing speed, reducing power consumption and improving energy efficiency. The research results of this paper provide an efficient processor scheme for data processing in power industry, and also provide new ideas and methods for the application of RISC-V architecture in power industry.

By analyzing the structural composition of Turbo codes, this paper delves into one of its crucial components, the interleaver, discussing its role, various types, and distinct characteristics. Given the interleaver's capability for composite interleaving, this study proposes a pseudo-random spiral interleaving method that balances the need for high randomness with ease of engineering implementation. The scheme is realized through programming in the hardware description language Verilog, and simulation results are presented. The findings validate that the pseudo-random spiral interleaver exhibits strong randomness and is relatively easy to implement in hardware, bearing significant engineering value.

The outreach of China's cultural and creative industries has been growing in recent years, playing a positive role in improving China's economic structure and promoting economic development that cannot be ignored. Meanwhile, China has paid more attention to the development of the cultural and creative product design industry, and has successively introduced several policies to support the cultural and creative product design industry, which has greatly enriched the types of cultural and creative products and services. In the construction of a knowledge graph for cultural and creative product design, domain characteristics need to be considered. This article focused on the characteristics and practical needs of cultural and creative product design knowledge, and designed a knowledge system to optimize knowledge extraction, fusion, storage, and supplement knowledge. A knowledge graph of cultural and creative product design was established to achieve understanding, retrieval, recommendation, Q&A, and decision-making assistance for cultural and creative product design knowledge. However, further research is needed to enrich relationship types, theoretical knowledge and multimodal knowledge, expand knowledge sources, improve quality evaluation and control, and strengthen the construction, sharing, and collaboration of sub domain knowledge graphs. The experimental results indicated that the semantic expression accuracy of the four aspects of clear definition of concepts, accurate description of relationships, semantic reasoning, and semantic linking in ordinary information retrieval systems was between 40% and 79%, which was not as high as the accuracy of knowledge graph systems

The power tower is an important part of the transmission grid. The quality of the inspection of a large amount of electric tower is an important factor in the safety of the power system. An electric pool image is constructed to realize the autonomous cruise of the UAV by three-dimensional point cloud reconstruction technology based on UAV aerial. This paper first uses the key point technology of UAV aerial image to collect important parts of the electric pole tower; then plans the cruise route based on the attributes and relative positions of different tower types and the current point position; the three-dimensional cloud and tower type templates are used to dynamically build a more accurate power tower image; finally, uses point cloud Data to dynamically plan the power tower path, and then realize UAV independent cruise, The accuracy of the method of building the method in the point cloud is verified by the construction of the actual electric tower line, and the three-dimensional reconstruction of the electric tower image is accurately realized.

This study proposes and implements a face detection and recognition system based on DL (Deep Learning) to address the challenges and requirements in the field of face detection and recognition. In terms of facial detection, modern DL models are adopted to improve the accuracy and speed of detection. In facial recognition, deep convolutional neural networks are used for feature extraction, and the accuracy of identity recognition is improved by introducing an angle robustness loss function. The experimental results indicate that the system has undergone multiple tests to verify its performance stability and practical application feasibility. The innovation of this study lies in fully utilizing DL technology to improve the performance of facial detection and recognition through training and optimization of large- scale datasets. By exploring different model architectures, loss functions, and training strategies, system performance was optimized and robustness was improved. In summary, this study provides strong support for further research and application in the field of facial detection and recognition, and provides powerful tools and methods for solving security and identity verification problems in the real world.

With the increasing demand for electricity and the wide application of new energy sources, the role of Artificial Intelligence (AI) in power grid construction management has gradually become prominent. By analyzing actual cases and current research results, this paper discusses the important role and challenges of AI in power grid construction management. In terms of role, AI shows its unique advantages through intelligent operation and maintenance management, equipment health monitoring and intelligent energy management. Intelligent operation and maintenance management system can monitor the status of power equipment in real time and predict potential faults, thus improving the operation efficiency of power grid. At the same time, the application of AI in energy management is expected to optimize the power production plan, better integrate distributed energy and energy storage systems, and promote the efficient use of renewable energy. However, AI faces a series of challenges in power grid construction management. Among them, data privacy and security are urgent problems to be solved, especially when sensitive data are involved in power system. The technical integration problem is also an important challenge. Due to the diversity of power equipment standards, the integration of AI system becomes more complicated. In addition, attention should be paid to social, economic and employment impacts, and relevant laws and regulations and ethical standards should be formulated to ensure the rational application of AI in power grid construction management. The research in this paper provides a combination of theory and practice for the development of AI in power grid construction management, and provides a reference for future policy formulation and technological innovation.

This paper presents a comprehensive exploration of the optimization of spare parts configuration schemes for ship equipment using a multi-agent simulation methodology. The focus lies in leveraging the failure rates of life-limited parts and employing advanced multi-agent modeling technologies to conduct simulated calculations pertaining to a ship's offshore operations. Through these simulations, the paper aims to ascertain the optimal outcomes for the spare parts configuration scheme. The research methodology involves the development and application of a multi-agent simulation method, specifically tailored to address the challenges associated with the mathematical modeling of complex systems. The proposed approach not only enhances the accuracy of spare parts configuration but also contributes to overcoming difficulties in modeling intricate systems. To substantiate the effectiveness of the proposed multi-agent simulation method, the paper includes illustrative examples demonstrating its application in real-world scenarios. The findings of the research highlight the method's efficacy in optimizing spare parts configuration schemes for ship equipment. This paper contributes to the field by presenting a sophisticated and practical approach to address the complexities inherent in mathematical modeling within the context of maritime operations. Based on the multi-agent simulation method, this paper primarily discusses the optimization of spare parts configuration schemes for ship equipment. By means of failure rates of life-limited parts as well as multi-agent modeling technologies, this method can be employed to perform the simulating calculation of a ship's offshore operations, thus determining the optimization outcomes of the spare parts configuration scheme. Research findings demonstrate that the multi-agent simulation method proposed in this paper addresses the difficulties related to the mathematical modeling of complex systems, with an example presented to illustrate the proposed method.

With good classification performance, digital medical image methods based on deep learning have received more and more attention. However, the problem of these methods is that the black-box characteristic of deep learning makes it difficult to explain the model and prediction results. In order to improve the interpretability and reliability of medical image classification tasks, this paper designs a network structure that introduces eliminating uncertainty in GCN feedback by using evidence theory and subjective logic. The diagnosis experiments of COVID-19 images show that the model can effectively improve the recognition accuracy of COVID-19 patients, and has good reliability.

China has a vast number of roads. With years of use and damage from severe weather and natural disasters, the service life of China’s roads is gradually being shortened. As a result, various road diseases, such as potholes, peeling, cracks, and more, will appear on the road surface. During the collection of road disease datasets, some exceptional cases are inevitable. For example, in low-illumination conditions or highly dark scenes, the visibility of the surrounding environment is very low, significantly reducing the image acquisition quality. Secondly, when the vehicle-mounted camera passes over uneven ground during the shooting process, the vehicle will experience severe jolts, causing the camera to shake irregularly. When the vehicle speed is unstable, the object motion will cause motion blur due to defocusing, which seriously degrades image quality and hurts subsequent advanced visual tasks such as object detection based on road dataset images. Hence, this study embraces a Retinex-based image optimization approach explicitly tailored to enhance the clarity of low-light, blurred road imagery. It has broad practical application prospects for road disease detection in low-illumination blurred environments.

As a precious heritage of ancient Chinese culture, the complex patterns and craftsmanship of Western Zhou bronze ware have always been of great concern to researchers and audiences. In this study, we first introduced the principle and characteristics of holographic projection technology, emphasizing its potential advantages in the field of digital display. Subsequently, we provided a detailed description of the research methodology and technical implementation, including digital scanning of cultural relics, 3D modeling, projection equipment, and design of interactive interfaces. We also conducted user research to collect feedback and demands from the audience regarding holographic projection displays. A study was conducted on the comparison between holographic projection technology and traditional textbooks in educational applications. The results showed that in terms of memory, holographic projection also scored seven, which shows that holographic projection technology is more helpful for students to remember what they have learned compared to the five points in traditional textbooks. These data indicate that holographic projection technology has potential applications in education, especially in improving students' interest, understanding, and memory, showing significant advantages compared to traditional textbooks.

Multimodal emotion analysis, blending machine learning and deep learning, is transforming computer-based human emotion recognition. This review examines the complexity of human emotions, generally classified into six types: joy, sadness, anger, fear, surprise, and disgust. It covers both unimodal and multimodal techniques, emphasizing their utility in fields like medicine, driving safety, and law enforcement. Unimodal technologies analyze single data types (text, image, audio), while multimodal fusion algorithms integrate these for heightened accuracy. Traditional machine learning methods, such as SVM and k-NN, are foundational for feature extraction and classification. Deep learning, using CNN and RNN architectures, advances emotion recognition by extracting intricate data patterns. Combining traditional machine learning with deep learning enhances precision, with the CNN-LSVM approach notably boosting classification accuracy. The paper focuses on three primary multimodal techniques: feature-level, decision-level, and hybrid fusion. Feature-level fusion merges various modality features to enrich classifier inputs. Decision-level fusion consolidates outcomes from different modalities, ensuring stable classification. Hybrid fusion, merging the prior two, offers a holistic approach for accurate emotion analysis. These methods, however, face challenges like increased computational demands and complex model training.

With the proposal of the strategic goal of "carbon peak, carbon neutral", the power supply side of China's power system is shifting from being dominated by thermal power units to being dominated by new energy units, and the intermittency and volatility of a high proportion of new energy sources have greatly increased the uncertainty of the power system, so that frequency, demand and ramping play an important role in guaranteeing the balance of supply and demand in the power grid, as active auxiliary services to satisfy the grid uncertainty. They play an important role in guaranteeing the balance between supply and demand of the power grid. Under the profound change of power system structure, it is necessary to further sort out the positioning, function and models of each auxiliary service, study the difference and connection between these three types of auxiliary services, and then clarify the difference between these three types of auxiliary services, taking into account the development and change of the power system. In addition, the impact of these three types of auxiliary services is analysed in the light of the changes in grid resources in the context of carbon peaking and carbon neutrality, and relevant suggestions are made for subsequent auxiliary service demand measurement.

With the advancement of the reconstruction of new distribution networks, new energy is continuously integrated into the power grid, and the utilization of power grid resources and production development are gradually becoming more complex and high-dimensional. Power supply service command needs to coordinate multiple application systems such as dispatch control, equipment management, and customer service, in order to improve the service quality of electricity demand for users. In response to this issue, this paper proposes a digital twin technology architecture for collaborative application of power supply service command. Due to the rapid updating of the status and location of resources involved in power supply service command coordination, and the need for accurate analysis of these rapidly changing resources in command coordination, relatively accurate decision-making conclusions can be given based on this. Therefore, digital twin technology is adopted to describe all resources involved in power supply service command and coordination, and to obtain the status recognition of various power grid resources through data collection and information transmission communication technology. Then, in the analysis layer of the digital twin model, fault analysis algorithms, power outage range analysis, power supply reliability analysis and other analysis algorithms are introduced to enhance the support effectiveness of the digital twin model for power supply service command coordination.

The goal of power supply service command is to solve the problems encountered by users during the electricity consumption process with high efficiency. Users often encounter equipment failures, load exceeding limits, voltage instability, and other issues during the electricity consumption process, and equipment failures are the most frequent events that occur during the electricity consumption process. The existing equipment maintenance process is analyzed by decision-makers from the power supply service command center based on the received user repair needs, and then work tasks are assigned to maintenance service personnel. However, this mode has the problem of excessive manual intervention and low efficiency. The article proposes an optimization method for equipment maintenance plan for power supply service command, analyzing the equipment maintenance plan from the perspectives of driving path of service personnel, allocation of repair resources, and vehicle availability coefficient. Take the importance of regional load, the ability of power supply service personnel, and the completeness rate of materials as input parameters for the design of emergency repair plans. Then design an evaluation function for the emergency repair plan, with the upper limit of the repair time as the constraint of the plan and solve for the optimal emergency repair plan. This method searches for multiple possible combination schemes among the resources required for existing equipment maintenance, and evaluates the feasibility of the schemes through evaluation models to optimize the existing equipment maintenance schemes and improve the efficiency of equipment maintenance.

With the increasing intelligence and flexibility of the power system, various operators have joined the models of the power market and grid dispatch to predict future electricity demand or power generation, providing data support for safe and stable operation of the power system and load demand prediction. However, due to the complexity and uncertainty of the power system, as well as the heterogeneity and distribution of operators, there are significant errors in load forecasting results. As for to improving the performance of federated learning models, it is an important and difficult problem to evaluate the value of multi-party electricity data. In this paper, we propose a method based on gradient ascent model unlearning to evaluate datasets in probabilistic load forecasting scenarios for various flexible resource operating entities in the power grid. By checking new model performance after removing each client, we evaluate the contribution of each client dataset to the global model. It can effectively measure the impact of each client dataset on the global model while protecting client privacy. We conducted experiments on different datasets to verify the effectiveness and feasibility of our method. Our experimental results reflect the adaptability and sensitivity of our method in four electric datasets.

In order to explore the feasibility and real-world application advantages of dynamic resource allocation in VoIP (Voice over Internet Protocol) networks using SIP (Session Initiation Protocol), this paper proposes a SIP-based approach to improve voice quality levels and optimize VoIP resource utilization. The method adopts HMM (Hidden Markov Model) to model the dynamic changes of SIP communication traffic and combines with the characteristics of SIP protocol to realize the dynamic adjustment of voice traffic resources within VoIP. The final realization provides good communication quality and network service QoS (Quality of Service) received by end-users through effective prediction and resource adjustment of SIP communication traffic and server load in VoIP communication, which has a constructive impact on improving the level of voice quality received by end-users.

Corpus linguistics studies the actual use of language in the real world, paying attention to the actual language use context and following the rules and laws of language use. Corpus linguistics serves as a rich repository of resources for foreign language education. Initially, printed texts like word dictionaries, grammar guides, and diverse teaching aids constituted vital components of the corpus. However, users cannot classify according to their actual needs, resulting in stereotyped and personalized resource databases that naturally reduce the due functions of the resource database. To address the limitations of collaborative filtering algorithms in English education applications, this paper emphasizes the importance of establishing an English instructional resource management system. This system can help improve the effectiveness of English education by providing teachers with a wealth of resources to choose from based on their specific needs and teaching goals. The management system should be able to categorize and label resources to ensure easy retrieval and utilization, as well as provide support for online education courses to improve their quality and accessibility.

This paper presents the advancements in image restoration technology using deep neural networks. This paper introduces the application of deep learning in image processing, including the development of deep neural network, the advantages of convolutional neural network in image recovery, and the challenges and future trends of deep learning in image recovery. The advancements in image restoration algorithms are explored, covering the image recovery algorithm based on sparse representation, the image recovery algorithm using the variational model (VBM), and super-resolution reconstruction techniques leveraging deep learning. Through experiment and analysis, we show the effectiveness of image restoration technology based on deep neural network and compare the performance of different algorithms in image recovery. The results show that deep learning methods have better performance in image recovery tasks.

This article proposes a panoramic mapping and interaction method for real-world images and power grid real-world twins. The data of the power grid scene is obtained through high-precision data acquisition technology, and accurate positioning is carried out. Secondly, the collected point cloud data is optimized and processed to construct a power grid real-world twin model, and texture mapping is performed. The panoramic mapping algorithm is used to map the power grid real-world twin model to the real-world image. Real time object detection and tracking, attitude estimation, and visual feedback and effect presentation are performed. The YOLO object detection algorithm and Kalman filter technology are used to achieve target detection, tracking, and state estimation. Finally, the Extended Kalman Filter (EKF) algorithm is used to reflect the changes in the real power grid, and it is inputted as an observation value. In EKF, through the prediction and update steps of EKF, Integrate the state of the model and sensor observations to perform incremental updates on the dynamic model.

The intelligent networked automobile industry is developing rapidly, however, its data security is also undergoing great challenges. In order to carry out the research on data security risk in the intelligent networked automobile industry, the current research status of data security risk in the intelligent networked automobile industry should first be comprehensively sorted out and analyzed, so as to find the entry point of the research. This study adopts the method of bibliometrics to statistically and analytically analyze the existing research on data security risk in the intelligent networked automobile industry from the perspectives of the number of existing literatures, relevance, and popular research directions, etc., so as to further sort out the content and progress of the research on data security risk in the intelligent networked automobile industry, and then provide certain theoretical guidance and practical suggestions for the healthy development of the intelligent networked automobile industry.

The transportation industry is accelerating its digital and intelligent transformation, and many regions continuously exploring the “smart” upgrade and transformation of toll stations. The application of target recognition technology in highway toll collection systems can effectively improve toll collection efficiency and accuracy. In view of the characteristics of monocular vision in existing recognition methods, this paper proposes a target recognition method for highway toll collection based on stereo vision, aiming to improve the accuracy and robustness of target recognition. Test results show that the designed method can achieve the recognition of human hand and high-speed ETC card or ETC payment code. Not only with high accuracy in target recognition, but also with good robustness, without the problem of a significant decrease in target recognition accuracy with an increase in the number of background target objects.

Epoxy resin as an excellent material is widely used in various power equipment. Therefore, for the safe and stable operation of power equipment, it is necessary to study the internal defects of epoxy resin. Although many studies have clarified the preparation method of defect samples and the fundamental phenomenon of deterioration caused by partial discharge (PD) occurring due to defects in inner epoxy, little is known about whether the defect model conforms to the reality. The relation between cavity size and discharge is not clear as well. Hence, this paper provides a realistic and convenient way to generate cavity defect samples by using a pipette gun. Partial discharge measurements with three different sensors have been used to understand the role of cavity size in the process of discharge in the cavity sample. Fast Fourier Transform is also used to study the relation between the frequency of discharge signal and cavity size. The result suggests that the size of cavity does have an effect on the discharge amplitude and the discharge frequency.

To integrate multispectral technology for high-precision spectral analysis of insulator pollution and aging, it is essential to develop a more accurate model for insulator detection to acquire precise position information of the insulators. This paper presents a novel insulator region detection model based on YOLOv5, which addresses the challenges of insulator multispectral image alignment and target detection. The model is designed to detect insulators on insulator multispectral images and extract their position information, thereby laying the foundation for evaluating the state of insulators using multispectral images.

Virtual Power Plants (VPPs) not only provide flexible regulation capabilities for the power system, but also gain profits by participating in the electricity market. VPPs purchase electricity in the market and sell electricity to aggregated resources. VPPs earn the difference between the purchase cost and sales revenue. The arbitrage principle of VPPs in the electricity spot market is analyzed. The difference between the purchase price and retail price provides the possibility for VPPs to gain additional revenue. A method of using prices to incentivize aggregated resources to adjust electricity consumption to achieve arbitrage is proposed. The example analysis proves the effectiveness of the proposed method. Both VPPs and aggregated resources can gain additional revenue.

In recent years, the utilization of Facial Expression Recognition (FER) has gained significant traction across diverse domains including healthcare, marketing, and human-computer interface. This trend has prompted a gradual shift towards the adoption of deep learning methodologies for facial expression recognition. The paper highlights the importance of FER systems in comprehending human communication, detailing traditional methods and recent advancements in the field. It proposes a new framework integrating attention mechanisms into the VGGNet model for FER, utilizing the FER-2013 dataset for evaluation. Data augmentation techniques are employed to enhance model robustness, and attention mechanisms are integrated into the VGGNet-SE model. Through Comparative experiments, it is shown that the model of VGGNet-SE-last which adds attention mechanisms after the last convolutional layer of VGG16 achieves the most significant improvement in accuracy and reduction in Top 5 error, particularly with a ratio of 16 in SENet. Experimental results indicate that after incorporating attention mechanisms, the model can more accurately focus on important regions related to emotions in the images, leading to improved performance in recognizing facial expressions. In the future, further research on FER will be implemented by suggesting further optimizations in model architecture and exploration of more complex datasets for future research in FER.

In view of the general goal of vocational education undergraduate in our school, relying on the platform of provincial electromechanical integration specialty group, a centralized and comprehensive practice system of electrical and electronics curriculum group is constructed, mainly to cultivate compound technical and technical talents for the region, the whole province and even the Yangtze River Delta region. Under the background of talent training requirements of "specialty and industry are connected, curriculum and post are linked", curriculum setting must take "skill application" as the starting point and pay attention to the cultivation of students' practical ability and innovation ability. However, the traditional basic electrical experiments are mainly based on book theoretical knowledge, which cannot give consideration to the cultivation of practical ability and innovative ability. Therefore, it is imperative to carry out reform. In the revision of the talent training scheme for electrical automation and mechatronics, combined with the current situation of basic electrical experiments, the idea of replacing basic electrical experiments with centralized comprehensive training for electrical and electronic engineering is put forward. After full discussion, it was finally decided that the major of mechatronics should give priority to the centralized comprehensive training of electrical and electronics instead of basic electrical experiments, and the major of electrical automation should carry out it the following year. The revision of the training scheme provides a policy guarantee for the reform of centralized comprehensive training of electrical and electronic engineering instead of basic electrical experiments. It aims to cultivate high-quality and high-level skilled talents who can understand both theory and practice and improve teachers' professional level and teaching efficiency.

Effectively integrating advanced information methods to promote the digital transformation of highways, improving their management efficiency and service level, is an important way to enhance the high-quality development of highway networks. Through the collection and integration of dynamic traffic information in the whole space and time, vehicle road collaboration can provide accurate, real-time and personalized travel information services. The multiple rounds of policy, technology, and market promotion have led to the rapid development of vehicle-road collaboration on intelligent highways, but the construction and application process is generally faced with the weak ability to reach the information, the sense of user experience is not strong and other issues. The coverage rate of roadside equipment and the penetration rate of vehicle end are increasing, laying the foundation for the construction of an integrated and refined travel information service model. This paper analyzes the key technologies and research status of information exchange and information services in vehicle road collaboration, looks forward to a new model of integrated travel services, and relies on the digital foundation of intelligent highways to create a highway information service model based on vehicle road collaboration technology. This provides a reference for expanding innovative applications of vehicle-road collaboration and enhancing the construction and operation efficiency of intelligent highways.

This study optimizes power grid transmission prices using machine learning algorithms and then proceeds to work on economic evaluations to explore its application potential. The operation data of a provincial power system from 2019 to 2023 were obtained in the study, and many machine learning algorithms, such as linear regression, decision tree, random forest, and support vector machine, were used for model building and analysis. The results indicate that the proposed machine learning-based pricing strategy outperforms the current company pricing strategy, realized particularly in a 10% increase in revenue, an 8.3% reduction in costs, an 83.3% growth in profit, and 8.2% and 2.1% up in user satisfaction and energy efficiency levels, respectively. The comprehensive evaluation scores of the optimization strategy are worked out through this multidimensional economic evaluation index system, using the hierarchical analysis method, to verify the enormous advantages in economic benefits and operation efficiency.

This research explores the application of smart grid extensive data analysis in electricity price mechanism design and economic evaluation. A thorough analysis of power load data, user behavior data, and meteorological data is conducted to establish an optimized electricity price mechanism. This mechanism is expected to enhance the economic benefits of the power system and ensure its stability. First, the study uses big data technology to achieve accurate collection and processing of multi-source data, providing detailed data support for the design of electricity price mechanisms. Secondly, through time series analysis and regression modeling, the impact of the electricity price mechanism on the electricity market and user behavior was assessed. The results showed that the dynamic electricity price mechanism can significantly improve economic benefits in different seasons and load conditions. Empirical studies further prove that the optimized electricity price mechanism is effective in flattening the fluctuation of load and improving the balance between supply and demand. This study concludes that big data analytics in smart grids can optimize the electricity price mechanism and significantly enhance the economic benefits of the power system, providing a scientific basis for the sustainable development of smart grids.

With the rapid development of deep learning, image segmentation techniques based on deep learning have been widely applied to biomedical image analysis. However, due to the high variability in the shape, size, and color of cell nuclei, as well as noise and blurring during image acquisition, achieving precise segmentation of nucleus images remains challenging. To address these issues, this article proposes a nucleus image segmentation method based on an attention mechanism and edge optimization. By incorporating an efficient shuffle attention mechanism, the method captures channel information while retaining spatial information. Using the Canny operator to precisely extract edge details, an additional parallel auxiliary branch is introduced to extract low-level semantic information. After feature fusion and further optimization, the segmentation edges' clarity and accuracy are enhanced. Experimental results show that the MaskR-CNN with the addition of shuffle attention mechanism and edge optimization, demonstrates significant advantages in various metrics, effectively enhancing the feature extraction and detail retention capabilities for nucleus segmentation, mitigating under-segmentation or over-segmentation issues, and improving the accuracy of nucleus segmentation.

Through image enhancement, the quality of SAR images can be improved to make the details and structures in the images clearer. Through image fusion, the information of different images can be synthesized, the accuracy of detection can be improved, and the probability of false detection and missing detection can be reduced. In this paper, gamma correction, filter denoising and structure enhancement are used to improve the quality and contrast of SAR images. which synthesizes the information of different images and improves the accuracy of change detection. Through experimental verification, the accuracy of the above method in the process of SAR image change detection is verified, in order to provide some support for the technical application in related fields.

Image classification is one of the important basic tasks in computer vision. As the basic module of various complex tasks, ResNet of various types plays an important role as a back- bone for various tasks. This paper aims to use a series of effective methods to improve the performance of traditional ResNet and, to some extent, improve its generalization ability. We improve the methods of ResNet in terms of refined image conversion, optimizer improvement, adaptive learning rate scheduling, difference classification weighting, and experimentally using validation set integration, as well as in the structure of ResNet itself by incorporating residual structures in the network by using ResNet. With these bar improvements, our new baseline performance has made a significant difference, and we also have excellent performance in generalization migration to multiple tasks.

Concrete cracks significantly threaten buildings' safety and service life, necessitating early detection and prevention. The low efficiency and high cost of traditional manual detection methods make obtaining high-quality detection data in complex environments difficult. This paper aims to introduce deep learning artificial neural network-based crack detection methods for concrete components, highlighting the achievements of various algorithmic models in image processing and their performance evaluation.