Fault diagnosis of metal materials in sensors of HVAC systems under non-steady state conditions
The operating status of HVAC systems can change over time.Diagnosing sensor metal material faults un-der such non-steady state conditions can lead to a decrease in diagnostic performance.Therefore,a method for diag-nosing sensor metal material faults in HVAC systems under non-steady state conditions is studied.Firstly,the col-lected data is preprocessed using CEEMD and wavelet transform methods to reduce noise interference under non-sta-tionary conditions.Then,fault features are extracted based on known types of sensor metal material faults.Finally,an LSTM network is used to achieve the diagnosis of metal material faults in HVAC system sensors.The experi-mental results show that the proposed method has good denoising effect,high detection accuracy,and high detection efficiency,and is suitable for fault diagnosis of metal materials in sensors in HVAC systems.
non-steady state conditionsHVAC system sensorslong short-term memory neural networkmetal material fault diagnosiswavelet transform method
NETL
NSTL
万方数据
