Aiming at nonstationary property of mechanical vibration signals,Empirical mode decomposition(EMD) can adaptively decompose signals to several intrinsic oscillation modes,and different mode contains different status information of an equipment.By means of approximate entropy(ApEn),the complexity of the original signal and its intrinsic oscillation modes can be quantified,thereby the intrinsic mode comlexity of mechanical vibration signals can be quantitatively evaluated. One can use this method to understand the complex dynamic behavior of rotors ,and monitor the change of running conditions of a mechanical equipment to prevent deterioration of fault.The engineering application results showed that this method can extract the fault information of a mechanical equipment ,and provide an effective technology for condition monitoring and fault diagnosis of a mechanical equipment.