Abstract: The traditional reliability assessment methods based on probability statistics depend on failure lifetime data which have to be obtained through life test and accelerated life test. It is not suitable to utilize this method for the real-time operating equipment. In such a case, the performance degradation data based on equipment condition can provide useful information about the reliability assessment for the equipment. By integrating traditional reliability modeling techniques with vibration signal feature extraction of equipment condition, a new method of reliability assessment based on equipment condition vibration feature by using proportional hazards model has been is proposed for the first time. This method can establish relationship between low-level information (such as condition parameters) and high-level information (like reliability statistics) to build relation. It is suitable for estimating reliability parameter of the equipment operation, and providing effective reliability assessment. Thereby providing technical support for the preventive maintenance of equipment based on condition with reliability as center. A practice example of locomotive wheel set bearing is given to demonstrate the rationality and effectiveness of this method.

Key words: Equipment condition, Proportional hazards model, Reliability assessment, Vibration feature