| With the rapid increase in the number of large public buildings in my country,the demand for energy continues to increase.Building energy consumption accounts for 39% of the total energy consumption,which has a certain impact on the sustainable development of society.Among them,building energy consumption is mainly the energy consumed by the HVAC system.Therefore,it is urgent to establish an accurate mathematical model of building rooms and optimize the control system of the HVAC system to achieve the goal of reducing the energy consumption of the HVAC system.The variable air volume air-conditioning system has been used in various large public buildings with its significant energy saving and comfort advantages.However,the multi-zone variable air volume air-conditioning system of large public buildings is a complex multi-variable and strongly coupled system,which urgently needs accurate mathematical models and advanced control strategies to achieve precise control of the room temperature in each zone.This paper first based on the RC network method,the energy balance theory is used to establish a single room temperature and humidity coupling model.The model is based on a complex building envelope,the coupling relationship between indoor temperature and humidity is removed through mathematical methods,and an accurate solution of the single-room temperature and humidity coupling model is given,humidity can be used as a separate variable.Through MATLAB simulation software,given the various parameters of the building envelope and indoor and outdoor environmental parameters,the dynamic changes of the nodes of the wall and the indoor temperature are simulated.Then the single room temperature and humidity coupling model is extended to the entire building,and the multi-zone room temperature and humidity coupling model is established,and the solution of the indoor temperature of each room in the multi-zone is obtained in the form of state space equations,and then the indoor temperature dynamics variety of the rooms in each zone are obtained.Secondly,this paper uses graph theory and fluid pipe network transmission and distribution theory to abstract and simplify the variable air volume air conditioning system,modeling the resistance characteristics of the main components of the variable air volume air conditioning system,and the analysis method of basic loop of networks(MKP)calculation method is used to solve the mathematical model of the pipe network of the variable air volume air-conditioning system to obtain the air flow through each room.Then the deep belief neural network is used to establish the hourly air supply prediction model of the variable air volume air conditioning system.Comparing the prediction results of this model with those of BP,Elman,and fuzzy neural networks,the results show that the deep belief neural network has the highest prediction accuracy,and the average absolute relative error,root mean square relative error,and determination coefficient are 1.555%,0.789% and0.9975,which shows that the model established in this paper can accurately and effectively predict the supply air volume of variable air volume air-conditioning.Finally,this paper combines fuzzy clustering algorithm with model predictive control,and proposes an indoor temperature predictive control model for multi-zone variable air volume air conditioning systems based on fuzzy clustering algorithm,and verifies the control effect of this model on a multi-zone VAV air conditioning test bench,at the same time,verify the accuracy of the multi-zone building model when the air volume changes greatly. |
