摘要
针对随机分布系统控制及干扰抑制问题,提出一种基于神经PID控制算法。利用神经网络的自适应和自学习能力在线调整链接权值,实现PID控制器三个参数的整定;同时,在建立状态反馈系统基础上,通过线性矩阵不等式(LMI)给出控制器的设计方法,并引入H∞范数,以提高系统的抗干扰能力,实现系统输出概率密度函数(PDF)追踪目标概率密度函数的控制。计算机仿真结果表明,该算法可实现系统输出概率密度函数追踪目标概率密度函数,抑制追踪控制中的干扰,并确保系统具有满意的鲁棒性和稳定性。
Aimed at the problems of random distributed system control and interference suppression,a neural PID control algorithm is proposed.The adaptive weight and self-learning ability of the neural network were used to adjust the link weights online to realize the tuning of the three parameters of the PID controller.At the same time,based on the establishment of the state feedback system,the design method of the controller was obtained by the linear matrix inequality(LMI).And the H∞norm was introduced to improve the anti-interference ability of the system,and realize the control of the output probability density function(PDF)tracking target PDF.The computer simulation results show that the proposed algorithm can achieve the system output PDF tracking target PDF,suppress interference in tracking control,and ensure that the system has satisfactory robustness and stability.
作者
屈毅
Qu Yi(Department of Electrionics and Information Engineering,Xianyang Vocational Technical College,Xianyang 712000,Shaanxi,China)
出处
《计算机应用与软件》
北大核心
2023年第4期275-281,328,共8页
Computer Applications and Software
基金
陕西省教育厅专项科学研究计划项目(17JK1169)
咸阳职业技术学院博士科研启动基金项目。
