Fault detection method for hydraulic turbine bearings of energy storage units based on multi-source information fusion
In response to the problem of insufficient accuracy in fault detection of hydraulic turbine bearings in energy storage u-nits,a fault detection method based on multi-source information fusion is proposed.Specifically,the NMD algorithm is improved and designed to improve the accuracy of decomposed signals and ensure the accuracy of fault diagnosis.In the improved NMD algorithm,the extraction termination conditions for main and sub harmonics have been improved,and the signal decomposition process has been adjusted to make its signal decomposition logic clearer;Extracting time-frequency ridges based on GMP fitting can effectively avoid the impact of noise on time-frequency ridge extraction and improve the accuracy of subsequent reconstructed signals.Through simula-tion experiments,it has been proven that the improved NMD algorithm has good signal denoising performance,and the reconstructed vibration signal has high accuracy without distortion,basically meeting the improvement requirements;Through physical experiments,it has been proven that the improved NMD algorithm can accurately extract vibration signals with high vibrations from raw signals con-taining noise and abnormal peak signals,and has certain application effects in fault detection of hydraulic turbine bearings in energy storage units.
万方数据
维普
