论文部分内容阅读
模糊神经网络控制器不依赖于被控对象精确的数学模型,又能根据被控对象参数的变化自适应调节控制规则和隶属函数参数,但是模糊神经网络控制器在线修正权值计算量大、过度修正权值还可能导致系统剧烈振荡.针对以上问题,提出了在线修正计算中仅对控制性能影响大的权值进行修正,以减小计算量;根据偏差及偏差变化率大小,基于TS模型自适应调节权值修正步长,抑制控制器输出的剧烈变化,避免系统发生振荡.仿真结果表明模糊神经网络控制器的优化设计方法可以改善系统控制性能.
Fuzzy neural network controller does not depend on the accurate mathematical model of the controlled object, but also can adjust the control rules and membership function parameters adaptively according to the change of the controlled object parameters. However, the fuzzy neural network controller has a large amount of online correction weights, In order to solve the above problems, this paper proposes that only the weights that have a great influence on the control performance should be corrected in order to reduce the amount of calculation. According to the deviation and the rate of change of deviation, Adjusts the steps of weight adjustment, and suppresses the drastic change of controller output to avoid oscillation of the system.The simulation results show that the optimal design method of fuzzy neural network controller can improve the system control performance.