单神经元PID控制仿真与实现研究  

Research on simulation and implementation of single neuron PID control

作　　者：鲁博 张寿明[2] 冯欣怡 郭英英 LU Bo;ZHANG Shouming;FENG Xinyi;GUO Yingying(Haojing College,Shaanxi University of Science and Technology,Shaanxi Xi'an 712046,China;Kunming University of Science and Technology,Yunnan Kunming 650093,China;Xi'an Institute of Technology,Shaanxi Xi'an 710032,China)

机构地区：[1]陕西科技大学镐京学院咸阳市人工智能重点实验室,陕西西安712046 [2]昆明理工大学信息工程与自动化学院,云南昆明650093 [3]西安工商学院计算机学院,陕西西安710032

出　　处：《工业仪表与自动化装置》2024年第1期3-8,共6页Industrial Instrumentation & Automation

基　　金：陕西省科研计划项目(No.21JK0554)。

摘　　要：经典PID控制参数整定方法繁琐并且容易对设备造成损伤,此外,当系统对象存在非线性、环境变化或者其他因素时,系统的稳定性和响应速度也会受到一定的影响。为了解决这些问题,该文引入神经网络与PID控制方法相结合,控制对象选择双容水箱,对系统进行多次仿真、对比,然后提出神经网络控制仿真参数设置方法,进而将神经网络PID控制算法应用于实际控制系统中,实验证明该创新方案不仅解决了PID控制方法存在的问题,还找到了一个低成本的升级方案,为控制系统提供了重要的技术支持。The classical PID control parameter tuning method is cumbersome and easy to cause damage to the equipment.In addition,when the system object has nonlinear,environmental changes or other factors,the stability and response speed of the system will be affected to some extent.In order to solve these problems,this paper introduces the combination of neural network and PID control method,the control object is selected two-capacity water tank,the system is simulated and compared several times,and then the neural network control simulation parameter setting method is proposed,and then the neural network PID control algorithm is applied to the actual control system.The experiment proves that this innovative scheme not only solves the problems of PID control method,but also finds a low-cost upgrade scheme,which provides important technical support for the control system.

关 键 词：PID控制 神经网络 智能 低成本 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]