Research on Junction Temperature Control Method of LED Array

At present, high-power LED arrays usually use active heat dissipation to reduce the LED junction temperature. In view of the fact that the temperature control system with active heat dissipation mode has the characteristics of time-varying uncertainty and nonlinearity, and the traditional PID is difficult to meet the control requirements, a fuzzy radial basis function neural network PID control method based on particle swarm optimization (PSO-RBF-PID) was proposed. The control model of high-power LED array was established, and the control methods of ordinary PID, Fuzzy-RBF and PSO-RBF-PID were used for MATLAB software simulation experiments, and the results showed that the fuzzy RBF neural network PID control method optimized by PSO algorithm effectively improved the robustness and adaptability of the system, shortened the control time, and had a good control effect on the LED, which had a certain feasibility.