基于BP神经网络的自适应补偿控制方法  

Adaptive Compensative Control Scheme Based on BP Neural Network

作　　者：陈忠华[1] 李雷[1] 赵力[1] 

机构地区：[1]辽宁工程技术大学电气与控制工程学院,辽宁葫芦岛125105

出　　处：《计算机仿真》2012年第7期202-205,共4页Computer Simulation

摘　　要：研究工业过程控制系统补偿问题,对于一类模型未知的SISO非线性系统,传统的控制方法不能获得被控系统的精确数学模型,因而在系统稳定性和鲁棒性上存在缺馅,控制效果不佳。为了提高被控非线性系统的稳定性和鲁棒性,提出了一种基于BP神经网络的自适应补偿控制方法。首先,通过逆系统理推导了被控系统输出和伪控制量之间的误差,然后误差进行在线自适应BP神经网络补偿,从而实现对被控系统的BP神经网络自适应补偿控制,且采用Lyapunov理论证明BP神经而网络的收敛性和闭环系统的稳定性。计算机仿真表明所提方法明显提高了非线性系统的鲁棒控制性能。As for a class of SISO nonlinear system with unknown mathematical model, stability and robustness cannot be guaranteed with traditional control schemes. In order to improve the control performance, an adaptive BP neural network compensate control scheme was proposed. Firstly, the error between the α-th derivative of the system output and the pseudo-control was deduced with the inverse system method, then an adaptive BP neural network was designed to compensate the error. The uniform ultimate boundness of the tracking error of the closed loop system and the weight estimation errors of BP neural network were proven with Lyapunov stability theory. And by computer simu- lation and comparison, it was concluded that the proposed control scheme can greatly improve the robust control per- formance of the nonlinear system.

关 键 词：逆系统 自适应补偿控制 非线性系统 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]