基于RBF在线辨识的PID整定

Self tuning PID controller of RBF based on-line identification neural network

针对非线性系统,提出一种新型的径向基函数(RBF)辨识网络的控制算法,根据RBF网络在线辨识被控对象的离散模型,得到被控对象Jacobian信息,利用BP网络在线自适应整定PID参数。通过RBF网络对系统在线辨识,克服不确定性对系统性能的不利影响,从而解决传统PID控制鲁棒性差及受精确数学模型限制的问题。通过实际算例验证,并与常规PID控制作对比,仿真结果表明,该控制算法有较强的自适应性和鲁棒性,其抗干扰和适应参数变化的能力都优于常规PID控制。

In view of nonlinear system, a new control algorithm of RBF identification network is proposed. First, RBF network identifies discrete model of plant on-line, which can get plant＇ s Jacobian information. Then, BP network on-line adaptively adjusts PID parameters. Using RBF network to on-line identify the system, which overcomes the adverse effect generated by the uncertainty, and the problem that the traditional PID control method has poor robustness and is limited to the accurate mathematical model is solved. The method is tested by practical example, and is compared to conventional PID control. Simulation results show that the control method has better adaptability and robustness, and has the advantages of higher antiinterference ability and adaptability to parameters＇ changing than conventional PID control.