摘要
针对静变电源模糊比例微分积分(模糊PID)控制隶属度和控制规则一旦确定后无法更改等不足,为提高静变电源输出电压的质量,提出以模糊神经比例微分积分(模糊神经PID)控制代替模糊PID控制。利用神经网络自学习的特点完善和调整模糊逻辑规则,并与PID稳态控制性能的优势相结合,实时地对静变电源电压控制量进行调整。仿真结果表明,模糊神经PID控制系统具有良好的鲁棒性和自适应能力,对干扰也有较好的抑制效果,解决了模糊PID控制器在静变电源控制中的精度不太高、自适应能力有限的难题,满足了静变电源输出的要求。
A fuzzy neural PID controller was designed to instead of the fuzzy PID controller to promote the qual- ity of the static inverter output voltage. The new method had the advantages of fuzzy logic and neural network, so that the reasoning of the system was speeded up, and the fuzzy control rules was also improved continually. The accuracy of the system control was advanced for it combining with the steady performance advantages of PID control which made the real-time measurement and concurrent adjustment to the system control. Based on the MATLAB/SIMULINK, the simulation results showed that, compared with the fuzzy PID control, the con- trol system had better performance of adaptive capacity, it solved the problems of low precision and adaptive ca- pacity, and it also met the requirements of high robust and rapid in the system.
出处
《探测与控制学报》
CSCD
北大核心
2013年第2期59-63,共5页
Journal of Detection & Control
