Application Research Of BP Neural Network In Automatic Control System Of Heat Exchange Station

Centralized heating is an indispensable part of people’s life in the northern part of China,and the heat exchange station,as an important part of centralized heating,is related to people’s quality of life in terms of its system operation.Secondary water supply of heat exchange station is the main way used in most residential areas,but there are still some problems in operation control,such as mismatch between heat supply and demand,high energy consumption and waste of resources.The working mode of traditional heat exchange station heating system can no longer meet the needs of modern social development,so the use of intelligent regulation and control means to improve the quality of secondary water supply is imperative.In this thesis,BP neural network PID control is applied to the automatic control system of heat exchange station,and the regulation and effect analysis are carried out.First of all,in order to cope with time lag,nonlinearity,uncertainty and other problems commonly found in heat exchange station heating temperature control systems,this thesis applies BP neural network PID control to the secondary water supply temperature control system,using the self-learning and self-adaptive capabilities of BP neural networks to optimize the PID control parameters and achieve an optimal control strategy for the secondary water supply temperature.Subsequently,the overall control scheme of the system is formulated with PLC as the control core,and the secondary water supply temperature control system is used as the focus of the study,and its mathematical model is established using the step response curve method,and the simulation model of the control system is built in Matlab.Secondly,based on the actual project requirements and the overall system architecture,the hardware design and software programming research of the automatic control system of the heat exchange station are completed.Select PLC and field equipment and complete the system I/O point address allocation and PLC control cabinet circuit design;After that,the setting of CPU parameters and the programming of the control system are completed in the PLC programming software,the programming of BP neural network algorithm is completed by using Matlab,and finally the human-computer interaction is realized by designing the touch screen.Finally,a university heat exchange station project in Zhangjiakou is used as the experimental platform for the commissioning of the system.Using OPC technology to realize data transfer between Matlab and PLC,and the BP neural network algorithm is applied to the actual control system.Through the step response experiment and step disturbance experiment on the secondary water supply temperature control system,it is found that BP neural network PID control is superior to conventional PID control,and its performance quality is more prominent,its overshoot can be controlled within5%,the adjustment time is reduced by 25% when the system is not disturbed,and when the system sudden disturbance phenomenon,it can quickly recover to the set value and always maintain near the set value,with better stability and robustness.