Study Of The Sensor Fault Diagnosis In Air Conditioning System Based On Principal Component Analysis

With the increasing demands on indoor environment quality(IEQ) and energy consumption in modern buildings, heating,ventilation,air-conditioning(HVAC ) systems become more and more complex. In order to reduce power consumption,maintain comfort in building,reduce wear on HVAC equipment, automation fault detection and diagnosis come into forth. In this dissertation, the fundamental theory of principal component analysis(PCA) have been deeply analyzed. A novel fault diagnosis method based on improved dynamic PCA has been presented, compared with conventional static PCA, it can applied into dynamic systems. 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 has been presented. In complicate systems, to construct a single PCA model including all the system variables hardly provides satisfactory performance in the fault detection and diagnosis application. To solve this problem, the measured variables are divided into several groups according to the principles governed the system, using data fusion knowledge to detect faults and combined with neural network to fulfill fault recognition. Specific issues that deserve more attention and deeper understanding have been presented. In the end, the proposed methods are verified by a simulation test for air handling unit of air conditioning system.