Sensor Fault Detection And Diagnosis In HVAC System

Faults in heating, ventilation and air-conditioning(HVAC)system have severe effect on the energy consumption and operation cost of the system , building indoor comfort and the wear of equipment. Timely detecting and eliminating faults in a HVAC system can save energy, reduce operation costs of system, improve indoor air quality(IAQ)and maintain indoor comfort. Therefore, automated fault detection and diagnostics is becoming more and more interested.. Accurate and reliable sensor measurement is the premise of exact automated fault detection and diagnosis in HVAC system. At the same time, it is also the important guarantee of accurate supervising and reliable control of HVAC system. So, it is significant to develop the strategies for the sensor fault detection and diagnosis of HVAC systems.In this paper, principal component analysis (PCA) approach is used to detect and diagnose sensor bias faults and drifting 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 minimizing the total unreconstructed variance in this paper.PCA approach divides the measurement space into principal component subspace (PCS) and residual subspace (RS). The normal data is included in the PCS and the fault or noise is included in the RS. So, fault can be detected by detecting the projection of the measurement data in RS. In this paper, squared prediction error(SPE), a statistic variable, is used to detect whether the system is faulty.The essence of fault reconstruction is a process of seeking an estimation value for correct value which corresponding to fault measurement data. Reconstruction via iteration is used for fault reconstruction in this paper. The essence of iterative reconstruction is a process of sliding the measure to PCS along the direction of fault.Fault identification is an important task of fault diagnosis. An index, SVI, is defined to identify faults in this paper. Research shows that SVI index can identify faults in temperature sensors and in fresh air flow sensor but it can't identify which is faulty when faults occurs in supply air flow sensor and return air flow sensor.