基于神经网络的摆动电机的模糊PID控制  

Fuzzy PID control of oscillating motor based on neural network

作　　者：霍建楠 王自强[1] HUO Jian-nan;WANG Zi-qiang(School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China)

机构地区：[1]北京航空航天大学自动化科学与电气工程学院,北京100191

出　　处：《电子设计工程》2018年第12期129-133,138,共6页Electronic Design Engineering

摘　　要：摆动电机是一种直流无刷的有限转角电机,它是特种电机的一种。它的执行机构能很好的适应环境,精度高,响应速度快。在非线性系统中,提出一种基于BP前馈网络模糊PID方法来提高系统稳定性和克服参数不确定性。利用BP神经网络对模糊PID控制器的参数和非线性进行调节补偿,从而减少模糊控制器对切换项的增益的需求。分析BP神经模糊PID中,BP神经网络,模糊算法和PID控制如何优化、互补和配合。最后用matlab仿真结果来说明BP前馈神经网络模糊PID控制器具有无超调、稳定性强、很好的抗干扰等优点,并非线性系统具有一定鲁棒性。The oscillating motor is a kind of Brushless DC motor with limited angle. The actuator can well adapt to the environment,with high accuracy and fast response. In nonlinear systems,fuzzy PID control based on BP feedforward network is proposed to improve system stability and to overcome parameter uncertainties. The BP neural network is used to compensate the parameters and nonlinearity of the fuzzy PID controller,so as to reduce the demand of the fuzzy controller for the gain of the switching term.Analyzed in BP neural fuzzy PID,BP neural network control,fuzzy control and PID control how to optimize with each other. Finally,the MATLAB simulation results show that the BP neural network fuzzy PID controller has the advantages of no overshoot,good stability,good anti-interference,and so on. It is not a linear system with certain robustness.

关 键 词：摆动电机 BP神经网络 模糊PID控制 MATLAB仿真 

分 类 号：TM359.6[电气工程—电机]