C.-C. Chiu, C.-Y. Liaw, and C.-M. Wu (Taiwan)
Hierarchical Bayesian Neural Network, Degradation Process, Failure Time, Markov Chain Monte Carlo
In today's technological world, most of us rely on the continued functioning of a wide array of complex machinery and equipment. We all expect our utilized facilities, such as computers and electrical appliances, to perform functionally well whenever we use them. It can be easily understood that the failure of these equipment could bring enormous trouble to us. However, the reliability for some devices with few or no failures in their life tests is very hard to access if a traditional life test which records only time-to-failure was utilized. To solve this problem, the analysis of the over time degradation processes is always considered in the practical cases. The realization of the degradation processes is expected to be represented by the constructed degradation model. Based on the developed models, the failure times for devices and the time-to-failure distribution can be estimated. In this paper, a hierarchical Bayesian neural network model (HBNN) with autocorrelated residuals is proposed to construct a broad class of degradation models. For findingtheappropriateestimatesofmodel’sparameters, the Markov Chain Monte Carlo (MCMC) algorithm is applied. A fatigue crack growth data is used as an example for illustrating the modeling procedure of HBNN. By specifying the coefficients in the HBNN, we successfully identify the heterogeneity. Moreover, the prediction intervals of future degradation processes for evaluating the prediction accuracy are provided..
Important Links:
Go Back
