Fault Detection And Diagnosis Of Refrigeration System Based On Multivariate Statistical Analysis

It is necessary to carry out fault detection and diagnosis(FDD) study of refrigeration system. Because various failures of cooling system occurred frequently, not only can cause a lot of energy waste, and the equipment safety, reliability, economy will decline. In this paper, models are made with multivariate statistical analysis (mainly principal component analysis and designated component analysis) and professional features. Then the models are used to do fault detection and diagnosis with the experimental data. The results showed that it performed well for both single and concurrent failures.First, analyzing cooling system and the common failures to learn the theoretical connection between signs and failures(symptoms and causes), providing a theoretical basis for the later research. Briefly introduced ASHRAE experiments and use the experimental data as historical data for fault detection and diagnosis studies. Introduced multivariate statistical analysis used in the paper, leading to the method of principal component analysis and designated component analysis.Secondly, models of principal component analysis were established with ASHRAE experimental data. Then the models were used to do fault detection and diagnosis for both single failure(only one fault occurs) and concurrent failures (two or more simultaneous failures) of refrigeration system. But the diagnosis results were not so satisfactory. Therefore, with the combination of principal component analysis and professionalism (theoretical knowledge and practical experience), designated component analysis(DCA) were established. First we established an observation space of a multi-variable system, which consists of normal subspace, fault subspace and residual subspace. Then observed data were shadowed to the fault subspace and normal subspace, establishing DCA space projection framework. The fault detection problem is converted into energy significant detection of observed data y(j) shadowing to the fault subspace, and fault diagnosis problem is then transformed into energy significant detection of observed data matrix Y shadowing to the subspace.At last, a dedicated fault detection and diagnosis test stand has been established by designing energy balance system and introducing fault simulation lines and components for and into an air-source heat pump of 16.8kW rated cooling capacity. Typical faults that could be simulated include refrigerant leakage/undercharge, overcharge, liquid line restriction, compressor valve leakage, reduced evaporator water flowrate, reduced condenser air flowrate, condenser fouling, thermal expansion valve over or less pre-tightened, etc. Experiments have been done for some types of the individual faults and two or three faults happening simultaneously. Variation of the critical parameters while faults happening was analyzed and possible cause or reasons were discussed. Then the models established before were used to do fault detection and diagnosis and they performed well.