Paper
5 March 2018 Optimization design of LED heat dissipation structure based on strip fins
Author Affiliations +
Proceedings Volume 10710, Young Scientists Forum 2017; 107102H (2018) https://doi.org/10.1117/12.2317711
Event: Young Scientists Forum 2017, 2017, Shanghai, China
Abstract
To solve the heat dissipation problem of LED, a radiator structure based on strip fins is designed and the method to optimize the structure parameters of strip fins is proposed in this paper. The combination of RBF neural networks and particle swarm optimization (PSO) algorithm is used for modeling and optimization respectively. During the experiment, the 150 datasets of LED junction temperature when structure parameters of number of strip fins, length, width and height of the fins have different values are obtained by ANSYS software. Then RBF neural network is applied to build the non-linear regression model and the parameters optimization of structure based on particle swarm optimization algorithm is performed with this model. The experimental results show that the lowest LED junction temperature reaches 43.88 degrees when the number of hidden layer nodes in RBF neural network is 10, the two learning factors in particle swarm optimization algorithm are 0.5, 0.5 respectively, the inertia factor is 1 and the maximum number of iterations is 100, and now the number of fins is 64, the distribution structure is 8*8, and the length, width and height of fins are 4.3mm, 4.48mm and 55.3mm respectively. To compare the modeling and optimization results, LED junction temperature at the optimized structure parameters was simulated and the result is 43.592°C which approximately equals to the optimal result. Compared with the ordinary plate-fin-type radiator structure whose temperature is 56.38°C, the structure greatly enhances heat dissipation performance of the structure.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lingyun Xue, Wenbin Wan, Qingguang Chen, Huanle Rao, and Ping Xu "Optimization design of LED heat dissipation structure based on strip fins", Proc. SPIE 10710, Young Scientists Forum 2017, 107102H (5 March 2018); https://doi.org/10.1117/12.2317711
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Light emitting diodes

Neural networks

Convection

Particle swarm optimization

Optimization (mathematics)

Evolutionary algorithms

Thermal effects

Back to Top