摘要：AbstractIn response to the global energy shortage problem, the energy-saving optimization of HVAC system is used to alleviate, and timely diagnosis and evaluation of faults can reduce unnecessary energy consumption. The difficulty in troubleshooting is that it is difficult to distinguish fluctuations in parameters caused by environmental changes from fluctuations in parameters of the chiller system. In addition, the coupling information of each internal device also causes difficulty in fault location. In this paper, using a hidden Markov process to obtain state parameters reduces the interference of the coupling information to the fault state of each component. In addition, an unscented Kalman filter is used to obtain the current state of the estimate, and then the type and extent of the fault is determined based on the percentage of the estimate fluctuation. Finally, SPC rules are used to obtain dynamic control limits to reduce the interference of the outdoor environment to the fault diagnosis system, thereby reducing the rate of misdiagnosis. The experimental results show that the method effectively diagnoses the chiller failure and prompts that the diagnosis is timely and accurate.
