RBF神经网络的PID控制研究与仿真

Study and Simulation on PID Control Based on RBF Neural Network

研究工业控制系统优化问题,由于工业控制领域中存在复杂非线性时变系统,很难确定精确模型。传统PID控制局限于线性系统,控制效果不理想。为了提高控制精度,提出一种RBF神经网络辨识的PID控制方法。首先利用RBF神经网络线辨识被控对象的离散模型,获得PID参数在线调整信息,然后利用单神经元对控制器参数进行在线自适应整定,从而实现系统的智能控制。通过实例进行验证,并与传统PID控制方法进行对比,结果表明控制方法具有响应速度快、控制精度高等优点,且具有较强的自适应性、鲁棒性和抗干扰能力,为控制系统设计提供了新方法。

Industrial control systems are complex nonlinear time-varying systems,while the traditional PID control is limited to linear system,therefore the control effect is not ideal.In order to improve the control precision,the paper proposes a control method based on RBF neural network.Firstly,discrete models is identified by RBFNN controller to get PID parameters tuning information,then single neuron controller is used to set the parameters so as to realize the intelligent control system.The proposed method is verified,the results show that the control method has faster response time,higher control precision compared with the traditional PID control methods,and is of strong adaptability,robustness and anti-interference ability.