摘要
该文从加载控制器角度出发将BP神经网络算法引入加载系统削弱多余力矩对系统的影响,提高加载精度。建立了电动伺服加载系统的数学模型,分析了多余力矩产生的原因以及基于结构不变原理存在的局限性。介绍了BP神经网络控制算法基本原理,并给出了具体控制结构及相应算法,设计了一种BP/PID复合控制器。仿真结果表明,复合控制器有效地抑制了系统的多余力矩,降低跟踪误差,改善加载系统的动态性能,提高了跟踪精度,增强了稳定性。
This paper introduces the intelligent control algorithm of BP neural network into the loading system to weaken the influence of surplus torque on the system, and to improve the accuracy of load- ing. The mathematical model of the electric servo loading system is established, and the reasons of the surplus torque and the limitation of the structure invariance principle are analyzed. This paper introduces the basic principle of BP neural network control algorithm, gives the concrete control structure and the corresponding algorithm, and designs a kind of BP/PID composite controller. The simulation result shows that the composite controller can not only effectively restrain the surplus torqu~ of system, reduce tracking error and improve the dynamic performance of the loading system, but also improve the tracking precision and enhace the stability.
出处
《工业仪表与自动化装置》
2017年第2期8-13,共6页
Industrial Instrumentation & Automation
关键词
电动加载系统
BP神经网络
多余力矩
复合控制
electric loading system
BP neural network
surplus torque
compound control