某火炮伺服系统的回声状态网络自抗扰控制

为解决火炮伺服系统面临的一系列非线性因素,设计一种基于回声状态网络的自抗扰控制(active disturbance rejection control,ADRC)策略.使用回声状态网络(echo state network,ESN)实现自抗扰控制重要参数的在线整定,并引入梯度下降算法与改进后的灰狼优化算法(grey wolf optimization,GWO)对回声状态网络进行训练.仿真结果表明:该新型控制方法能有效提高火炮伺服系统的动态响应性能、抗干扰性能以及随动跟踪精度,满足火炮伺服系统所要求的性能指标.

Echo State Network Active Disturbance Rejection Controlof Certain Type Gun Servo System

In order to solve a series of nonlinear factors faced by gun servo system, an active disturbance rejectioncontrol (ADRC) strategy based on echo state network (ESN) is designed. The echo state network is used to realize theonline tuning of the important parameters of ADRC, and the gradient descent algorithm and the improved gray wolfoptimization (GWO) algorithm are introduced to train the echo state network. The simulation results show that the newcontrol method can effectively improve the dynamic response performance, anti-interference performance and trackingaccuracy of the gun servo system, and meet the performance requirements of the gun servo system.