基于DRNN的非线性模型预测控制研究  

Research of Nonlinear Model Predictive Control Based on DRNN

作　　者：李军[1] 张宇[2] 王纪森[1] 

机构地区：[1]西北工业大学自动化学院,陕西西安710072 [2]中国飞行试验研究院,陕西西安710089

出　　处：《计算机仿真》2010年第8期9-13,共5页Computer Simulation

摘　　要：针对飞机液压系统某地面试验装置具有非线性、慢时变的特征,常规的控制算法难于实现精确控制。为了提高系统的实时性和精度,提出了基于DRNN神经网络的非线性模型预测控制算法。控制算法应用对角递归神经网络DRNN作为非线性系统的预测模型,同时采用了具有全局优化能力的启发式遗传算法作为滚动优化工具。将这一控制算法进行仿真试验,仿真试验结果表明,基于DRNN的NMPC对于装置具有自适应能力,控制精度较传统的PID控制有明显的提高。It is difficult to realize accurate control to the field test instrument of airplane hydraulic system using the general control algorithm because of its time-varying and nonlinear feature.Therefore,a kind of nonlinear model predictive control algorithm based on diagonal recurrent neural network（DRNN） is proposed.The control algorithm uses DRNN as the nonlinear predict model,what is more,the heuristic genetic algorithm acts as its rolling optimization tool.The nonlinear model predictive control algorithm has been applied to simulation of the field test instrument,the simulation result shows that the nonlinear model predictive control algorithm based on DRNN is an adaptive control algorithm on the field test instrument and the control accuracy realized by NMPC is much higher than by the traditional PID controller.

关 键 词：非线性系统 对角递归神经网络 启发式遗传算法 非线性模型预测控制 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]