Online Automatic Fault Detetion And Diagnosis On Primary Return Air-Conditionging Systems Based On Expert Rules

A fault in air-conditioning system will result in several serious consequences,such as the deterioration of the indoor environment,the high energy consumption,the increase in operational costs and the equipment wear inteusity,reduction in the life span equipment,and so on.Nowadays,air-conditioning systems and their control systems become much larger and more complex,which makes the faults difficult to be found. Thus faults are often covered by self-compensatory measures of the system so that the system keeps running sickly.It may cause the fault to deteriorate further,consuming more energy.The automatic fault detection and diagnosis techniques based on expert rules,simulating a field expert,can monitor the real-time operational parameters of system,find out the reason of the faults and suggest the possible faults.This method is especially applicable to a central air-conditioning system,as it is hard to establish a mathematical model of the system with indispeusible automatic control,while its operational conditions are complicated and deteriorate often occurres in its running process.It is a good way to ensure the system to run stably and energy efficiently.Firstly,based on faults frequently occurred in different operational modes of a primary return air-conditioning system in different seasons through a year.28 pieces of expert rules associated with control performance of each operational mode to conduct fault detection and diagnosis were put forward.Then,the software of HVACSim~+ was applied to simulate various fatitious false operational states of a HVAC system seveing an office building.Results indicate this set of expert rules proposed are feasible and effective for fault detection and diagnosis.And then,a software was developed to fulfill the fault diagnosis procedure based on sequential expert rule algorithms by using the Delph 6.0.It can transfer the real-time operational data of a air-conditioning system form the Access2003 database.And the operational data can be averaged over time to prevent the software from falling into fault alarm when the system is still in the adjusting process.Finally,the operational data obtained from HVACSim~+ simulation on a fatitious false operation state was written into a database.The software was used to transfer the operational data,compute,reason out the practical failure.It can give reasonable solutions to the faults.This indicates the set of expert rules can be used to carry out fault detection and diagnosis.