A Novel Reliability Assessment Method Based on Virtual Sample Generation and Failure Physical Model

An important issue in the validation of high-reliability and high-cost mechanical parts is reliability assessment for small sample, by means of tests. Since the sample should be as few as possible and the tests should be as short as possible, we combine some statistical approaches and failure physical models to estimate reliability for products subjected to degradation mechanism. In order to improve the accuracy of the reliability assessment of small sample, firstly, generate virtual samples based on the failure physical model and virtual sample generation method, only wear failure is studied in this paper. Then, a performance degradation model based on Normal process is used to assess the reliability. The effectiveness of the reliability assessment proposed in this paper is tested with a simulation, and an actual case about a machine tool guideway wear testing is studied.