Research On Fault Detection And Diagnosis Strategy Of VAV Air-Conditioning System

Variable Air Volume(VAV)air-conditioning system is widely used in high-end large and medium-sized office buildings.The structure and the control strategy of the VAV airconditioning system are more complicated compared with the constant volume air-conditioning system,thus the fault rate of the VAV air-conditioning system is higher than that of the constant volume air-conditioning system.The fault of the air-conditioning system will affect the quality and the comfort of the indoor air,and even lead to the increase of the energy consumption.Therefore,it is necessary to study the fault detection and the diagnosis technology of the VAV air conditioning system,predict the fault and detect the fault location in time,determine the severity of the fault,and take corresponding measures to reduce the equipment loss and the maintenance costs,so that the air-conditioning system can operate efficiently.The main research object of this paper is the air handling unit(AHU)of the VAV airconditioning system.The experimental research and the simulation research of the fault detection and the diagnosis technology of the VAV air-conditioning system were carried out.This paper summarized the common fault types of the AHU;using the experimental platform to simulate the five system fault conditions and normal system conditions,and use data mining as data samples.Through the determination of the training samples,the number of hidden layers and the number of neuron nodes,the fault diagnosis model of BP neural network can successfully identify the five types of AHU faults and verify the reliability of the method.Aiming at the temperature and the flow control of the air conditioning area,a fault mode library based on qualitative knowledge and quantitative analysis was established.The unknown modes were identified using six failure modes and verified by the TRNSYS simulation.Experimental and simulation studies show that the fault detection and the diagnosis method of the VAV air-conditioning system based on the BP neural network and the fault pattern recognition was effective and feasible,and it could successfully detect the fault type and cause of the VAV air-conditioning system,and can effectively guide operation and maintenance personnel to discover and repair fault in time.