Abstract:Wheel motor is the power source of a four-wheel electric vehicle independent drive system, its operating status is directly related to the safety of vehicle. In order to realize the monitoring of the running status of the hub motor, this paper proposes a new method--order self-separation method to adapt to its complicated driving conditions. This method overcomes the limitation of the traditional tracking method, the traditional method needs to collect the speed signal and the vibration signal at the same time, but it only researches the speed signal, and extracts the abnormal fluctuation components contained in the speed signal, and carries on the angular area resampling and the fourier transform to the fluctuating components, which by reference the way how to deal with unstable signal in terms of domain in traditional order analysis. It highlights the fault characteristics hidden in the speed signal, and then achieve the wheel motor fault diagnosis. Combined with Matlab simulation and wheel motor fault test, the results show that the self-separation method can effectively identify the fault information hidden in the speed signal.