摘要
针对传统的PID控制器控制效果欠佳以及分数阶PI^(λ)D^(μ)控制器参数复杂难以整定的问题,设计了一种基于误差反向传播(Back propagation,BP)神经网络算法的分数阶PI^(λ)D^(μ)控制器。首先,将分数阶PI^(λ)D^(μ)控制器数字化,然后通过BP神经网络算法调节突触权值,经调整后的输出量作为分数阶PI^(λ)D^(μ)控制器的参数值,最后分别采用分数阶和整数阶作为被控对象进行实验仿真,仿真结果证明了神经网络分数阶PI^(λ)D^(μ)控制器比传统PID控制器的具有超调量小、上升时间快、稳定性好的优点。
To solve the problems of poor control effect of traditional PID controller and the complex parameters of fractional order PI^(λ)D^(μ) controller,a fractional order PI^(λ)D^(μ) controller based on back propagation(BP)neural network algorithm is designed.Firstly,the fractional order PI^(λ)D^(μ)controller is digitized,and the synaptic weight of neural network is adjusted by BP algorithm.The adjusted output of neural network is taken as the parameter of fractional order PI^(λ)D^(μ) controller.Finally,the fractional and integer order are respectively used as the controlled object.In the experiment simulation,the simulation results prove that the neural network fractional order PI^(λ)D^(μ) controller has the advantages of small overshoot,fast rise time and better stability than the traditional PID controller.
作者
谢玲玲
秦龙
Xie Lingling;Qin Long(School of Electrical Engineering,Guangxi University,Nanning 530004,China)
出处
《南京理工大学学报》
CAS
CSCD
北大核心
2021年第4期515-520,共6页
Journal of Nanjing University of Science and Technology
基金
国家自然科学基金(61863003,61561007)
广西自然科学基金(2019GXNSFAA245019)。
作者简介
谢玲玲(1980-),女,博士,副教授,主要研究方向:电力电子的分析与控制,E-mail:xielingling@gxu.edu.cn。
