基于GA-PIDNN的液压弯辊控制系统设计

来源期刊:中南大学学报(自然科学版)2014年第11期

论文作者:张秀玲 徐腾 赵亮 樊红敏 臧佳音

文章页码:3800 - 3805

关键词:液压弯辊;PID神经网络(PIDNN);遗传算法;BP算法

Key words:hydraulic roll bending; PID neural network (PIDNN); genetic algorithm; BP algorithm

摘    要:针对液压弯辊控制系统的时变性、非线性和不确定性等特点,设计利用GA(遗传算法)优化的PID神经网络(PIDNN)液压弯辊控制系统。PIDNN控制器不仅具有不依赖被控对象数学模型的优点,而且有很好的动态性能,结构简单易于设计。利用GA代替BP算法对PIDNN权值进行优化,克服了BP算法易陷于局部极小的不足。2种优化方法的仿真结果对比表明:GA-PIDNN控制器能够使液压弯辊力快速达到目标值,并且具有较强的抗干扰能力。

Abstract: According to the characteristics of time-varying, nonlinear, and the uncertainty of the hydraulic roll bending control system, a hydraulic roll bending control system based on PIDNN (PID neural network) optimized by GA (genetic algorithm) was proposed. PIDNN controller has the advantage of simple structure and good dynamic performance, which is easy to design and does not rely on the mathematical model of controlled object. In order to overcome BP algorithm’s shortage of easy trapped in local minimum, GA was used to optimize the weights of PIDNN instead of BP algorithm. The comparative simulation results of these two optimization methods demonstrate that GA-PIDNN controller can make hydraulic roll bending force quickly reach the target and has a good ability of anti-disturbance.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号