摘要
本文针对热源塔热泵系统进行了分析,指出了热源塔热泵系统在各个部位之间存在高度耦合性,在进行控制设计时存在一定的困难。针对目前还没有热源塔热泵系统控制系统的研究,本文提出了在预测控制上加入优化后的输出的前馈反应的控制方法。介绍了该系统原理于组成,建立了压缩机,换热器,膨胀阀和热源塔的数学模型。为了在解决优化输出问题上提高计算速度,使用BP神经网络模型代替原有的数学模型,并且利用优化的输出量进行预测控制仿真。结果表明:在模型预测控制系统的基础上添加一个优化后的前馈反应的控制系统在控制过程中有着反应速度快、超调量小且能提升系统性能系数的特点。
This paper analyzes the heat source tower heat pump system,points out that the heat source tower heat pump system has a high degree of coupling between various parts,and there are certain difficulties in the control design.In view of the fact that there is no research on the control system of the heat source tower heat pump system,this paper proposes a control method that adds an optimized output feedforward reaction to predictive control.The principle and composition of the system are introduced,and the mathematical models of compressors,heat exchangers,expansion valves and heat source towers are established.In order to improve the calculation speed in solving the optimized output problem,the BP neural network model is used instead of the original mathematical model,and the optimized output is used for predictive control simulation.The results show that:Adding an optimized feedforward response control system on the basis of the model predictive control system has the characteristics of fast response speed,small overshoot,and the ability to improve the system performance coefficient during the control process.
作者
邴一伟
张小松
BING Yi-wei;ZHANG Xiao-song(School of Energy and Environment,Southeast University)
出处
《建筑热能通风空调》
2021年第10期7-12,共6页
Building Energy & Environment
基金
“十三五”国家科技支撑项目(No.2016YFC0700305)。
关键词
热源塔
神经网络
优化
预测控制
heat source tower
neural network
optimization
predictive control