Formalization of reliability model for assessment and prognosis using proactive monitoring mechanism

For purpose of estimating machine operational reliability and predicting the future state of reliability accurately and effectively, a reliability modelling framework is proposed by integrating prior distribution of failure with vibration signal feature extraction based on machine condition for proactive health management and reliability maintenance. The core idea of the proposed framework is not only utilize historical fault data of equipment operation and major repair, but also select appropriate dynamic condition monitoring signal. The feature evaluation is introduced to analyze the dynamic signals. The degradation index with accurate definition and apparent trend variation is extracted, and introduced to proportional hazards model, which effectively capture the effects of the covariates. Covariates could be dependent or independent of time and any quantity or combination of quantities that describe a system´s condition can be included in reliability models. Maximum likelihood is chosen to estimate the reliability parameters in proportional hazards model. This method permits in-situ assessment of machine reliability, and the evaluation can really reflect the trend of fault occurrence and development. It is suitable for realizing individual equipment reliability assessment and prognosis based on operational condition feature and performance degradation signal. A practice example of rolling bearing test device is given to provide operational data, and the reliability size and changing trend are analyzed in detail under different operational conditions and damaged severity.