MOMEDA Model Based Variational Mode Decomposition for Pelton Wheel Fault Detection

The identification of fault frequency is a vital step in fault diagnosis. For this purpose, variational mode decomposition (VMD) is being used widely as it has ability to represent the signal in Time-frequency domain. The VMD, on the other hand, often fails to analyse non-stationary signals with low-frequency noises. In this paper, a multipoint optimal minimum entropy deconvolution adjusted (MOMEDA) in combination with VMD is applied to enhance the performance of defect detection in the presence of low-frequency noises. Study has been made on bucket defects in Pelton wheel. Initially, the raw vibration signal is decomposed using VMD into set of modes. Second, the MOMEDA model is used to purify each mode by reducing low-frequency noise interference. A purified signal is reconstructed from the modes processed by MOMEDA. In the end, Hilbert envelop spectrum is extracted, which demodulates the purified signal and helps in determining the fault frequency. The same is verified from the theoretical faulty feature frequency of the turbine bucket. Performance of the proposed method in extracting fault frequency accurately is also compared with other two models: a) MOMEDA model with ensemble empirical mode decomposition (EEMD) and b) MOMEDA model with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN). The comparison results proved the efficacy and superiority of the proposed method.