论文部分内容阅读
对多变量系统设计了神经网络ＰＩＤ控制，解决了多变量系统Ｐ、Ｉ、Ｄ参数难以整定的问题。基于对象模型，提出了一种新型神经网络控制器的训练方法，该方法用多步二次型性能指标函数去训练控制器的权值，从而提高了控制器参数的收敛速度和系统的响应性能，降低了各通道之间的耦合。理论分析和仿真实验表明了该方法的有效性
Neural network PID control is designed for multivariable system, which solves the problem that it is difficult to set parameters of P, I and D in multivariable system. Based on the object model, a new neural network controller training method is proposed, which uses the multi-step quadratic performance index function to train the weight of the controller, so as to improve the convergence speed of the controller parameters and the response performance of the system , Reducing the coupling between the channels. Theoretical analysis and simulation experiments show the effectiveness of this method