Fuzzy control research of magnetic powder clutch based on variable learning factor particle swarm optimization

For the issue that applying traditional method of quantization factor fuzzy control for magnetic powder clutch in the engaging process during vehicle starts can not achieve optimization, a method of using variable learning factor partial swarm to optimize quantization factor of fuzzy controller has been put forward, the optimized quantization factor changes with the environmental condition as well as load and it can track the parameters´ change of fuzzy controller in real time, making the robustness and control accuracy of the fuzzy controller improved. The simulation results show that in the engaging process of magnetic powder clutch during vehicle starts, variable learning factor particle swarm optimization fuzzy control algorithm can effectively reduce the maximum collision during vehicle starts and the sliding friction work in the process of clutch engagement comparing with the traditional fuzzy control algorithm.