A Bayesian sequential testing method is proposed to evaluate system reliability index with reliability growth during development.The method develops a reliability growth model of repairable systems for failure censore...A Bayesian sequential testing method is proposed to evaluate system reliability index with reliability growth during development.The method develops a reliability growth model of repairable systems for failure censored test,and figures out the approach to determine the prior distribution of the system failure rate by applying the reliability growth model to incorporate the multistage test data collected from system development.Furthermore,the procedure for the Bayesian sequential testing is derived for the failure rate of the exponential life system,which enables the decision to terminate or continue development test.Finally,a numerical example is given to illustrate the efficiency of the proposed model and procedure.展开更多
Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting str...Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting strip.Navier-Stokes equation in fluid mechanics and stream function were introduced to analyze the rheological property of liquid zone and mushy zone,and deduce the analytic equation of unit compression stress distribution.The traditional hot rolling model was still used in the solid zone.Neural networks based on feedforward training algorithm in Bayesian regularization were introduced to build model for kiss point position.The results show that calculation accuracy for verification data of 94.67% is in the range of ±7.0%,which indicates that the predicting accuracy of this model is very high.展开更多
Real time remaining useful life(RUL) prediction based on condition monitoring is an essential part in condition based maintenance(CBM). In the current methods about the real time RUL prediction of the nonlinear degrad...Real time remaining useful life(RUL) prediction based on condition monitoring is an essential part in condition based maintenance(CBM). In the current methods about the real time RUL prediction of the nonlinear degradation process, the measurement error is not considered and forecasting uncertainty is large. Therefore, an approximate analytical RUL distribution in a closed-form of a nonlinear Wiener based degradation process with measurement errors was proposed. The maximum likelihood estimation approach was used to estimate the unknown fixed parameters in the proposed model. When the newly observed data are available, the random parameter is updated by the Bayesian method to make the estimation adapt to the item's individual characteristic and reduce the uncertainty of the estimation. The simulation results show that considering measurement errors in the degradation process can significantly improve the accuracy of real time RUL prediction.展开更多
Many Bayesian learning approaches to the multi-layer perceptron （MLP） parameter optimization have been proposed such as the extended Kalman filter （EKF）. This paper uses the unscented Kalman particle filter （UPF...Many Bayesian learning approaches to the multi-layer perceptron （MLP） parameter optimization have been proposed such as the extended Kalman filter （EKF）. This paper uses the unscented Kalman particle filter （UPF） to train the MLP in a self- organizing state space （SOSS） model. This involves forming augmented state vectors consisting of all parameters （the weights of the MLP） and outputs. The UPF is used to sequentially update the true system states and high dimensional parameters that are inherent to the SOSS moder for the MLP simultaneously. Simulation results show that the new method performs better than traditional optimization methods.展开更多
A systematic method was proposed to estimate the occurrence probability of defective piles(OPDP) from a site according to quality assurance inspection. The OPDP was firstly suggested as the criterion to weight the per...A systematic method was proposed to estimate the occurrence probability of defective piles(OPDP) from a site according to quality assurance inspection. The OPDP was firstly suggested as the criterion to weight the performance of a pile foundation. Its prior distribution and updating distribution were deduced to follow Beta distributions. To calibrate the OPDP, a dynamic estimation model was established according to the relationships between prior mean and variance and updating mean and variance. Finally, a reliability-control method dealing with uncertainties arising from quality assurance inspection was formalized to judge whether all the bored piles from a site can be accepted. It is exemplified that the OPDP can be substantially improved when more definite prior information and sampling formation become available. For the example studied herein, the Bayesian estimator of updating variance for OPDP is reduced from 0.0037 to 0.0014 for the first inspection, from 0.0014 to 0.0009 for the second inspection, and with less uncertainty by incorporating experience information.展开更多
基金supported by the National Natural Science Foundation of China (70571083)the Research Fund for the Doctoral Program of Higher Education of China (20094307110013)
文摘A Bayesian sequential testing method is proposed to evaluate system reliability index with reliability growth during development.The method develops a reliability growth model of repairable systems for failure censored test,and figures out the approach to determine the prior distribution of the system failure rate by applying the reliability growth model to incorporate the multistage test data collected from system development.Furthermore,the procedure for the Bayesian sequential testing is derived for the failure rate of the exponential life system,which enables the decision to terminate or continue development test.Finally,a numerical example is given to illustrate the efficiency of the proposed model and procedure.
基金Project(2004CB619108) supported by National Basic Research Program of China
文摘Rolling force for strip casting of 1Cr17 ferritic stainless steel was predicted using theoretical model and artificial intelligence.Solution zone was classified into two parts by kiss point position during casting strip.Navier-Stokes equation in fluid mechanics and stream function were introduced to analyze the rheological property of liquid zone and mushy zone,and deduce the analytic equation of unit compression stress distribution.The traditional hot rolling model was still used in the solid zone.Neural networks based on feedforward training algorithm in Bayesian regularization were introduced to build model for kiss point position.The results show that calculation accuracy for verification data of 94.67% is in the range of ±7.0%,which indicates that the predicting accuracy of this model is very high.
基金Projects(51475462,61374138,61370031)supported by the National Natural Science Foundation of China
文摘Real time remaining useful life(RUL) prediction based on condition monitoring is an essential part in condition based maintenance(CBM). In the current methods about the real time RUL prediction of the nonlinear degradation process, the measurement error is not considered and forecasting uncertainty is large. Therefore, an approximate analytical RUL distribution in a closed-form of a nonlinear Wiener based degradation process with measurement errors was proposed. The maximum likelihood estimation approach was used to estimate the unknown fixed parameters in the proposed model. When the newly observed data are available, the random parameter is updated by the Bayesian method to make the estimation adapt to the item's individual characteristic and reduce the uncertainty of the estimation. The simulation results show that considering measurement errors in the degradation process can significantly improve the accuracy of real time RUL prediction.
基金supported by the National Natural Science Foundation of China(7092100160574058)+1 种基金the Key International Cooperation Programs of Hunan Provincial Science & Technology Department (2009WK2009)the General Program of Hunan Provincial Education Department(11C0023)
文摘Many Bayesian learning approaches to the multi-layer perceptron （MLP） parameter optimization have been proposed such as the extended Kalman filter （EKF）. This paper uses the unscented Kalman particle filter （UPF） to train the MLP in a self- organizing state space （SOSS） model. This involves forming augmented state vectors consisting of all parameters （the weights of the MLP） and outputs. The UPF is used to sequentially update the true system states and high dimensional parameters that are inherent to the SOSS moder for the MLP simultaneously. Simulation results show that the new method performs better than traditional optimization methods.
基金Project(51278216) supported by the National Natural Science Foundation of ChinaProject(2013BS010) supported by Henan University of Technology Fund for High-level Talent,China
文摘A systematic method was proposed to estimate the occurrence probability of defective piles(OPDP) from a site according to quality assurance inspection. The OPDP was firstly suggested as the criterion to weight the performance of a pile foundation. Its prior distribution and updating distribution were deduced to follow Beta distributions. To calibrate the OPDP, a dynamic estimation model was established according to the relationships between prior mean and variance and updating mean and variance. Finally, a reliability-control method dealing with uncertainties arising from quality assurance inspection was formalized to judge whether all the bored piles from a site can be accepted. It is exemplified that the OPDP can be substantially improved when more definite prior information and sampling formation become available. For the example studied herein, the Bayesian estimator of updating variance for OPDP is reduced from 0.0037 to 0.0014 for the first inspection, from 0.0014 to 0.0009 for the second inspection, and with less uncertainty by incorporating experience information.
