Approach Research Of Sensor Fault Detection And Diagnosis In HVAC System

With HVAC systems'extensive applications and great improvements, the control systems of which have become more and more complicated. Business offices, industry factories, and even civil residential, that are all in increasingly demands of better stability, comfortable and energy saving of HVAC systems. It's beyond operators' ability to immediately detect and diagnose all faults in HVAC systems, which makes FDD (Fault Detection and Diagnosis) systems become more and more important whatever at present time or in the future.The fault diagnosis method based on multivariate statistical process control is an important branch of this field. While in recent years, the principal component analysis (PCA) method of different forms has been widely used and achieved good results.In this thesis, principal component analysis (PCA) approach is used to detect and diagnose sensor faults in variable air volume (VAV) system. PCA approach is used to model the system by measurement data in normal operation condition. So it is not necessary to build an analytic model of the system directly. Choosing the number of principal component is crucial when building a system model by PCA. The optimal number of principal component is decided by contribution rate of principal act in this paper. Conventional PCA can not be used in the nonlinear systems, in order to overcome this deficiency, a novel fault diagnosis method based on Kernel principle component analysis (KPCA) and neural network predictor is presented. At the same time, to improve the performance of neural network predictor, this thesis presents that Genetic algorithm is applied to optimize the weights of neural network, and achieved good results.In the thesis, algorithm programs are designed by using MATLAB. Use the algorithms above to diagnose sensor fault in Air-conditioning units of intelligent building, and testify this network until it has satisfying fault diagnosis ability. The subject is very meaningful to perfect sensor fault diagnosis methods in Air-conditioning system.Finally, the summary of research about this paper and opinions on fault detection and diagnosis research in HVAC system in future are given.