Research On Fault Detection Method Of AHU In HVAC System

Heating,Ventilation and Air Conditioning(HVAC)system is one of important subsystems in modern building system.The main function of HVAC system is known to control the temperature to make building environment more suitable for human.In HVAC system,air handing unit(AHU)is an important system.The function of AHU is to adjust the temperature and air supply in building according to human needs.As a subsystem of HVAC system,AHU needs to work continuously.Excessive workload,equipment wear and influence of natural environment are all reasons for AHU faults.When the fault cannot be timely detected and repaired,the system running in the fault state will generate additional energy consumption and cause more serious faults.Timely and accurate fault detection of AHU system can solve the problems mentioned above.In the process of AHU fault detection,it takes a lot of work to establish an accurate mathematical model based on the physical principle of AHU system,and modeling with its historical data will make the model relying too much on the historical process.It is difficult to establish a suitable mathematical model for the AHU fault detection by these two methods.At the same time,because AHU needs to interact with the outdoor environment,the human activities and the changes of the outdoor environment will cause the fluctuation of state parameters.However,the existing fault detection methods are difficult to distinguish these two kind of parameter fluctuation,and the reliability of fault detection results is difficult to be guaranteed.In order to solve the above difficulties,this paper develop a simplified grey box model and an AHU fault detection method based on the Ministry of Housing and Urban-Rural Development Science and Technology Project "Research on the Decision-Making Method of Green Operation Management of Existing Buildings Oriented to Energy Conservation and Livability"(2018-K1-019)and Shenyang Jianzhu University Hanyu Plan Project "Research on Energy Conservation Optimization Method of Energy Consumption Equipment in Sino-German Building Based on Lagrange Relaxation Frame"(XKHY2-64).The main work is as follows:Firstly,according to the physical principle of AHU in HVAC system,this paper establishes a simplified grey box mathematical model to describe the system state by using the model parameters,and reduce the workload of modeling.Secondly,the improved unscented Kalman filter is used to obtain the working state of the AHU.In the early stage of the system state estimation,the improved moving average method is used to optimize the initial settings of the unscented Kalman filter,which reduces the interference of the initial error on the state estimation of the unscented Kalman filter.In the process of state estimation,the modified factor is used to adjust the system state estimation to reduce the influence of model and working environment,and improve the accuracy of fault detection methods.Finally,dynamic control limits are set for fault detection to reduce the interference of outdoor environment on fault detection and the misdiagnosis rate.The data validation results of the AHU simulation model show that the fault detection method can effectively detect the faults in AHU system.