一种新型模糊神经网络滑模控制器设计  

Design of a New Fuzzy Neural Network Sliding Mode Controller

作　　者：焦帅 张斌[1] 韩旭[1] 王一凤 刘雪梅 王春禹 JIAO Shuai;ZHANG Bin;HAN Xu;WANG Yifeng;LIU Xuemei;WANG Chunyu(Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所,长春130033 [2]中国科学院大学,北京100049

出　　处：《组合机床与自动化加工技术》2023年第10期64-68,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金项目(12133009)。

摘　　要：针对基于PID算法的永磁同步电机矢量控制系统易受系统不确定性影响(负载扰动、模型参数变化)的问题,提出一种新型模糊神经网络滑模控制器。首先,在传统滑模控制的基础上,利用模糊神经网络控制器(FNN)构成滑模控制器(SMC)的切换控制项来完成对系统不确定性因素进行逼近以及对滑模控制切换增益的调节;其次,利用麻雀搜索算法来实时更新模糊神经网络滑模控制器参数,为了加快SSA的收敛速度以及防止其陷入局部最优,利用分数阶微积分对传统麻雀算法进行改进;最后,进行了仿真验证,表明所设计的新型模糊神经网络滑模控制器具有较强的鲁棒性和动态性能。Aiming at the problem that permanent magnet synchronous motor vector control system based on PID algorithm is vulnerable to system uncertainties(load disturbance,model parameter change),a new fuzzy neural network sliding mode controller is proposed.Firstly,on the basis of traditional sliding mode control,fuzzy neural network controller(FNN)is used to form the switching control term of sliding mode controller(SMC)to complete the approximation of system uncertainty factors and the adjustment of switching gain of sliding mode control.Secondly,the sparrow search algorithm is used to update the parameters of the fuzzy neural network sliding mode controller in real time.In order to accelebrate the convergence speed of SSA and prevent it form falling into local optimum,the traditional sparrow algorithm is improved by fractional calculus.Finally,the simulation results show that the designed new fuzzy neural network sliding mode controller has strong robustness and dynamic performance.

关 键 词：麻雀搜索算法 永磁同步电机 分数阶 模糊神经网络 滑模控制 

分 类 号：TH166[机械工程—机械制造及自动化] TG502[金属学及工艺—金属切削加工及机床]