Fault diagnosis of EHA-VPVM based on improved HHT

The system composition of the electro-hydrostatic actuator with variable pump displacement and variable motor speed(EHA-VPVM) was introduced. According to early fault signal features of EHA-VPVM, a new time-frequency analysis method Hilbert-Huang transform(HHT) was used. Two improved methods were introduced to solve problems of HHT, these problems were mode confusion and false intrinsic mode functions(IMF). The ensemble empirical mode decomposition(EEMD) was used to solve the mode confusion, and the correlation coefficient method was used to identify the effective IMF. In the early fault diagnosis experiment of EHA-VPVM prototype, Hilbert marginal spectrum and HHT spectrum were used to analyze the vibration signals of the permanent magnet brushless DC motor in different work conditions. The experiment results show the improved HHT well solved the problems of mode confusion and false IMF, and extracted the early fault features of the rolling bearing accurately. Moreover, the improved HHT method has good effect on the early fault diagnosis of EHA-VPVM.