基于粒子群优化BP神经网络PID的供热控制系统仿真研究  

Simulation research on back propagation neural network PID optimization heating control system based on partical swarm optimization

作　　者：李远航 高晓红[1] 姜庆龙 韩云峥 LI Yuan-hang;GAO Xiao-hong;JIANG Qing-long;HANG Yun-zheng(School of electrical and computer science,Jilin Jianzhu university,Changchun 130118,China)

机构地区：[1]吉林建筑大学电气与计算机学院,长春130118

出　　处：《吉林建筑大学学报》2024年第1期72-78,共7页Journal of Jilin Jianzhu University

基　　金：吉林省科技厅科技发展计划项目(20190303114SF).

摘　　要：供热系统技术属于清洁技术,但其能耗非常大,因此在供热系统中能源的损耗问题就显得尤为重要。与此同时,我国供暖过程多数是用传统PID对供暖系统进行控制,由于传统PID控制响应时间长、超调量高且受外界影响较大,造成能源未充分利用、浪费现象严重。因此针对此问题,提出了在供暖系统中采用一种基于粒子群优化BP神经网络PID的控制策略,不仅可以解决供暖时水温不稳定、水温上升时间长等问题,而且可以更好地解决能源未充分利用问题。本文建立供热系统的数学模型,然后利用Matlab中的Simulink设计并仿真粒子群BP神经网络PID控制器。实验结果表明,改进后的PID控制器抗干扰能力强且具有较好的鲁棒性,对供热控制系统有更好的控制效果。Heating system technology belongs to clean technology,but its energy consumption is very large,so the problem of energy loss in heating system is very important.At the same time,the heating system is controlled by traditional PID in our country.Because of the long response time,high overvoltage and it is affected by outside influence greatly,the traditional PID control results in insufficient energy utilization and serious energy waste.Therefore,aiming at this problem,a control strategy based on BP neural network PID optimization based on PSO is proposed in the heating system.It can not only solve the problems of unstable water temperature and long time of water temperature rise during heating,but also better solve the problem of underutilization of energy.First,a mathematical model of the heating system is established.Then the particle swarm BP neural network PID controller is designed and simulated by using Simulink in Matlab.The final experiment and results show that the improved PID controller has strong anti-jamming ability and good robustness,and has better control effect on the heating control system.

关 键 词：供热系统 粒子群 粒子群BP神经网络PID MATLAB 

分 类 号：TB273.4[一般工业技术—工程设计测绘]