基于改进QPSO算法的电动汽车模糊控制器参数优化

t present, electric vehicles often use DC brushless motor (BLDCM) as the driver, but the traditional method is difficult to design quantization factor and proportion factor of fuzzy controller in the BLDCM speed control system. An improved QPSO algorithm (AMF-QPSO) is proposed for the on-line tuning of the quantization factor and proportion factor. The AMF-QPSO algorithm takes the contraction-expansion coefficient (CE) control method as the research focus, proposes the particle activity concept, and uses it as the feedback quantity to realize the dynamic adaptive adjustment CE coefficient. At the same time, in order to prevent the high aggregation of the population, the elite group random cross learning mechanism is used to disturb some of the elite particles with low activity and enhance the diversity of the population in the later period. Finally, through the LabVIEW experimental platform, the performance of the AMF-QPSO algorithm is verified by a specific case. The experimental results show that the fuzzy PID controller optimized by AMF-QPSO has better control and adaptability than the standard fuzzy PID controller and the fuzzy PID controller optimized by QPSO.