A Reliability Analysis Method of Accelerated Performance Degradation Based on Bayesian Strategy

In order to achieve a longer life cycle, higher reliability is necessary for products. Generally, the traditional reliability analysis methods can be performed based on sufficient failure data. However, it is difficult to get such amount of failure data in practical engineering, which brings challenges to the traditional reliability analysis methods. Consequently, the traditional reliability analysis methods are not suitable for the case of no failure data or less failure data. In order to tackle the above challenges, an uncertainty analysis strategy using accelerated performance degradation information is given. While, in this method, the utilization of acceleration factor increases the number of model parameters, which lead to the loss of the accuracy of the model parameter estimation under the finite degradation data. To enhance the above strategy, a reliability analysis method of accelerated performance degradation based on Bayesian strategy is proposed in this study. The accelerated performance degradation analysis method combining historical degradation data and empirical information is introduced here. An engineering example of CNC machine tool function milling head is also used to illustrate the effectiveness of the given method.