基于神经网络算法的电加热过程温度控制  

TEMPERATURE CONTROL OF ELECTRIC HEATING PROCESS BASED ON NEURAL NETWORK ALGORITHM

作　　者：段东江 李海英 张贵杰 Duan Dongjiang;Li Haiying;Zhang Guijie(HBIS Group Tangsteel Company,Tangshan 063016,Hebei;School of Metallurgy and Energy,North China University of Science and Technology,Tangshan 063210,Hebei)

机构地区：[1]河钢集团唐钢公司,河北唐山063016 [2]华北理工大学冶金与能源学院,河北唐山063210

出　　处：《河北冶金》2023年第8期29-34,共6页Hebei Metallurgy

摘　　要：采用单神经元PID与传统PID控制相互结合的方法,探索高效电加热过程智能温度控制系统,并利用Simulink软件对传统PID控制和单神经元PID控制下的加热炉温度控制系统进行了仿真对比实验。结果表明,单神经元PID模型控制下的加热炉温度控制系统能够实时调整控制参数,使得控制参数和温度控制系统较为匹配,可以实现系统在最稳定的状态运行,该系统对干扰的抵抗性更好,鲁棒性更强。而实际试验中,单神经元PID控制下的加热炉温度控制系统在温度响应阶段具有较高的稳定性,同时在稳态阶段也拥有较高的控制精度,但是在初始温升过程中的响应速度较低。The intelligent temperature control system of high-efficiency electric heating process was explored by combining single-neuron PID and traditional PID control,and Simulink software was used to simulate and compare the temperature control system of heating furnace under traditional PID control and single-neuron PID control.The experimental results show that the temperature control system of the heating furnace under the control of the single-neuron PID model can adjust the control parameters in real time,so that the control parameters and the temperature control system are more matched,and the system can operate in the most stable state,and the system has better resistance to interference and stronger robustness.In the actual experiment,the temperature control system of the furnace under the control of single-neuron PID has high stability in the temperature response stage,and has high control accuracy in the steady-state stage,but the response speed is low during the initial temperature rise.

关 键 词：单神经元 PID 电加热炉 温度控制 鲁棒性 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]