基于神经网络和聚类算法的某破障武器自抗扰控制研究  

Research on active disturbance rejection control of a blast-breaking weapon based on neural network and clustering algorithm

作　　者：王嘉良 侯润民[1] 徐强[1] 龚永昌 钱雅婷 WANG Jialiang;HOU runmin;XU qiang;GONG yongchang;QIAN yating(School of MechanicalEngineering,Nanjing University of Science and Technology,Nanjing 210000,China)

机构地区：[1]南京理工大学机械工程学院,南京210000

出　　处：《自动化与仪器仪表》2024年第6期6-9,14,共5页Automation & Instrumentation

摘　　要：针对破障车其上破障武器发射时抗干扰性差的问题,提出了基于RBF和聚类算法的自抗扰控制方法。根据破障武器的随动系统原理,建立了相应的数学模型。同时为了解决自抗扰控制器内部参数繁多且较难整定的问题,采用了神经网络(RBF)对扩张观测器(ESO)和非线性控制率(NLSEF)参数在线整定的方法,并使用改进型聚类算法对RBF神经网络参数在线校正以达到更好的控制效果。通过Simulink仿真实验,证明该方法可以提高破障武器炮控系统的鲁棒性和响应速度,增强了系统的抗干扰能力。In order to solve the problem of poor anti-interference when the obstacle breaking weapon is launched,an active disturbance rejection control method based on RBF and clustering algorithm is proposed.According to the principle of the following system of the barrier breaking weapon,the corresponding mathematical model is established.At the same time,in order to solve the problem that the internal parameters of auto-disturbance rejection controller are too many and difficult to adjust,neural network(RBF)is used to adjust the parameters of extended observer(ESO)and nonlinear control rate(NLSEF)online,and improved clustering algorithm is used to adjust the parameters of RBF neural network online to achieve better control effect.Simulink simulation results show that the proposed method can improve the robustness and response speed of the gun control system,and enhance the anti-jamming ability of the system.

关 键 词：自抗扰控制器 扩张观测器 非线性控制率 神经网络 聚类算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]