Methods on reliability estimation and mechanism consistency test for accelerated degradation data

In this paper, we consider the application of the statistical degradation models for reliability analysis in non-destructive testing. Such models enable to estimate the reliability function (the dependence of non-failure probability on time) for the fixed critical level using the information of the degradation paths of tested items. The most widely used models are the gamma and Wiener degradation models, in which the gamma or normal distributions are assumed as the distribution of degradation increments, respectively. Using the computer simulation technique, we have analysed the accuracy of the reliability estimates, obtained for considered models. The number of increments can be enlarged by increasing the sample size (the number of tested items) or by increasing the frequency of measuring degradation. It has been shown, that the sample size has a greater influence on the accuracy of the reliability estimates in comparison with the measuring frequency. Moreover, it has been shown that another important factor, influencing the accuracy of reliability estimation, is the duration of observing degradation process.

In this paper, the Jeffreys priors for the step-stress partially accelerated life test with Type-II adaptive progressive hybrid censoring scheme data are considered. Given a density function family satisfied certain regularity conditions, the Fisher information matrix and Jeffreys priors are obtained. Taking the Weibull distribution as an example, the Jeffreys priors, posterior analysis and its permissibility are discussed. The results, which present that how the accelerated stress levels, censored size, hybrid censoring time and stress change time etc. affect the Jeffreys priors, are obtained. In addition, a theorem which shows there exists a relationship between single observation and multi observations for permissible priors is proved. Finally, using Metroplis with in Gibbs sampling algorithm, these factors are confirmed by computing the frequentist coverage probabilities.

In this paper, we propose a practical method to estimate the milling head reliability based on the accelerated degradation test and Bayesian lifetime analysis method. By carefully analyzing the failure mechanism of the milling head, we adopt the accelerated degradation test to identify and parameterize the failure law of this component. The equivalent lifetime for the Bayesian lifetime analysis is also derived from the accelerated degradation data. Finally, the estimation is carried out based on the parametric failure law and the equivalent lifetime using the Bayesian Weibull lifetime analysis method. The reliability estimation of the milling head can be improved by the combination of the failure mechanical based accelerated degradation test and the Bayesian field lifetime data analysis. The proposed method is a practical and cost-effective reliability estimation method for highly reliable and expensive mechanical component with a pre-defined failure mechanism.

Existing joint maintenance decision research typically ignores remaining useful lifetime (RUL) predictions for the accelerated degradation of equipment. A joint maintenance decision method for the replacement and spare-parts ordering strategy based on RUL prediction using accelerated degradation data for equipment is proposed in this paper. First, an RUL prediction model under accelerated stress is built by considering the proportional relationship between the drift coefficient and diffusion coefficient in the Wiener process. Second, based on the principle of step-by-step estimation, accelerated degradation test (ADT) data of the equipment are used to estimate the a priori unknown parameters. Finally, based on the RUL prediction results, a joint optimization model for the replacement and spare-parts ordering strategy is developed. Through example verification and cost parameter sensitivity analysis, the proposed method is shown to effectively improve the accuracy of RUL prediction and the scientific value of the joint optimization plan for equipment replacement and spare-part ordering, which is important to many engineering applications.

To estimate the quality of the induced predictive model we generally use measures of averaged prediction accuracy, such as the relative mean squared error on test data. Such evaluation fails to provide local information about reliability of individual predictions, which can be important in risk-sensitive fields (medicine, finance, industry etc.). Related work presented several ways for computing individual prediction reliability estimates for single-target regression models, but has not considered their use with multi-target regression models that predict a vector of independent target variables. In this paper we adapt the existing single-target reliability estimates to multi-target models. In this way we try to design reliability estimates, which can estimate the prediction errors without knowing true prediction errors, for multi-target regression algorithms, as well. We approach this in two ways: by aggregating reliability estimates for individual target components, and by generalizing the existing reliability estimates to higher number of dimensions. The results revealed favorable performance of the reliability estimates that are based on bagging variance and local cross-validation approaches. The results are consistent with the related work in single-target reliability estimates and provide a support for multi-target decision making.

This paper describes reliability assessment from measured data of double-stress accelerated degradation test which taking functional unit as research object. Most of performance degradation reliability analyses pay attention to components. Thereby assessments are relatively easy because they have simple function and clear failure mechanism. However, in reality, functional units compose system, and the reliability of a functional unit is affected by all of the jointing components synthetically which puts up a unitary degradation process. On this consideration, double-stress uniformly orthogonal accelerated test is putting on the whole board of functional units under temperature and relative humidity stresses (80 °C,55%), (100 °C,25%), (90 °C,85%), (110 °C,65%). Accelerated test design is introduced in this paper and the collected measured degradation data are shown by figures. According to statistical analysis reliability information is inferred from test data and assessment result is compared to results educed by other methods. Conclusion is result gain by this present way and is basically consistent with other studies. Functional unit reliability assessment based on performance degradation data is feasible and reliable.

High quality in systems and products is a wide term which include : reliability, initial performance, easy to use, security and compatibility under different environments. The most important characteristic of quality is reliability because of its impact in productivity, maintenance and costs. This paper shows the failure distribution estimation at component level for a non repairable item(considering only first failure) using Matlab. Failure distribution is important for reliability prediction and optimal maintenance policy. Although reliability prediction is important in the early stages of the design is also important in studies of maintenance requirements and reliability growth. The failure data for mechanical components sample was recorded and then the statistical analysis was done using Minitab, fitting the best statistical model between Normal, Lognormal, Exponencial and Weibull distribution. The minimum Anderson Darling coeficient was found for Weibull Distribution. Because the Minitab program assume the location parameter in zero which can deal to a wrong conclusion, a program for nonlinear programming was written in Matlab to find the parameters for Weibull distribution. The three parameters founded were â=3.6 , c=3284.5, a=1140.8. The common functions in reliability engineering were generated: failure density function, reliability function, mean time between failures, and failure rate function. The reliability estimation at any given stage as well as prediction of future reliability can be accomplished once the failure distribution function has been calculated . The statistical models not only represent the system , but enable the manager to assess progress and evaluate the quality of a product at a given point in time.

Risk-based group decision making regarding renewable energy schemes using a stochastic graphical matrix model

There are many difficulties in statistical analysis of multi-stress accelerated life testing, such as establishing the accelerated model and solving pluralism likelihood equations. With a focus on these difficulties, the Grey-SVM based life and reliability prediction method for multi-stress accelerated life testing is proposed, with the accelerated stress level and the reliability as SVM inputs, and the corresponding Grey AGO processing failure data as outputs. Simulation and case study shows that the method has high prediction accuracy and with less amount of training samples than neural network.

This research examines the nonlinear tracking problem where a single moving observer passively monitors noisy bearings emitted from a source and processes this data to obtain estimates of source position and velocity. Specific optimal paths are derived for general scenarios characterized by the effective range-to-baseline ratio. Although the work addresses bearings-only estimation, the theoretical findings are applicable to a large class of optimization problems. This is illustrated by extending the analysis to include localization with a more general angle-of-arrival measurement. Observability requirements are the first consideration with regard to observer motion. A simplified criterion for establishing such requirements is presented. This criterion is then utilized to determine necessary and sufficient conditions for a unique tracking solution in a class of nonlinear problems involving both bearing and other angle-of-arrival measurements. Emphasis is then directed toward establishing a theoretical relationship between system observability, observer motion, and estimation accuracy. This is accomplished by analyzing properties of the error covariance matrix and Fisher Information matrix. The index of optimality is the Cramer Rao lower bound. As is common with nonlinear optimization problems, an explicit analytical solution for optimal observer motion is difficult to obtain. The formation of the estimation problems considered here leads to a cost function that is not amenable to conventional study. However, it is shown in this work how decomposition of the Fisher Information matrix leads to a numerical procedure that generates optimal (and suboptimal) observer motion comprised of a series of constant course trajectories. In the limit, these trajectories converge to the continuous solution. Observer trajectories are also derived analytically for bearings-only estimation by utilizing an alternative criterion. Optimality is defined here as those paths that maximize this criterion. The problem is solved within the context of classical control theory by invoking necessary conditions involving the Hamiltonian. The error ellipses associated with the trajectories are compared and analyzed. It is shown that observer motion involves a trade-off between increasing bearing-rate and decreasing range. Particular characteristics of an observer path and its effect on estimation accuracy depend on the range-to-baseline ratio of the scenario initially encountered. (Abstract shortened with permission of author.)

Prediction of remaining useful life using the field monitored performance data provides a more realistic estimate of life and helps develop a better asset management plan. The field performance can be monitored (indirectly) by observing the degradation of the quality characteristics of a product. This paper considers the gamma process to model the degradation behavior of the product characteristics. An integrated Bayesian approach is proposed to estimate the remaining useful life that considers accelerated degradation data to model degradation behavior first. The proposed approach also considers interaction effects in a multi-stress scenario impacting the degradation process. To reduces the computational complexity, posterior distributions are estimated using the MCMC simulation technique. The proposed method has been demonstrated with an LED case example and results show the superiority of Bayesian-based RUL estimation.

In reliability analysis, engineers lacks confidence to verify the mean time to failure of high power light emitting diodes from aging or degradation tests. The lifetime information, however, is important to manufacturers to improve the quality of light emitting diodes. The paper provides a bootstrap computation procedure to infer the mean time to failure of high power light emitting diodes. Assume that the degradation paths collected from an accelerated degradation test follow a Wiener process, confidence interval of the mean time to failure of high power light emitting diodes is found based on bootstrap percentiles. An illustrating example of GaN-based LEDs is used to demonstrate the use of the proposed method.

The construction of interval estimates for system reliability from component test data has presented statisticians and reliability engineers with several challenging problems. This article outlines the difficulties in making such estimates and presents a selection from the many diverse and sometimes ingenious statistical approaches that have been made. The paper summarizes some recent developments and indicates directions for some further research.

In this paper it was demonstrated that by applying the classical way of reliability lifetime prediction, the reliability of a product can no longer be guaranteed in some cases and for some failure mechanisms. This is caused by reduced reliability margins under the influence of increasing fields, current and power densities on the one hand, by introduction of new materials and devices on the other hand, and by ever increasing failure rate requirements imposed by the market. As a result, the reliability engineers have started to rethink the classical reliability assessment methodology. By taking into account the changes in failure behavior and statistics under influence of scaling trends, and analyzing the impact of such failures on the device and circuit operation, new reliability margins can be gained. This will, however, require more interaction between technology, reliability and design engineers, in order to define realistic failure specifications and new chip failure criteria for each type of circuit. In this paper some elements of such a new reliability assessment methodology have been demonstrated by using oxide breakdown as the failure mechanism for a case study where the classical reliability margins are already reduced to zero. It is expected, however, that for most of the other failure mechanisms a similar trend will emerge, and similar solutions, requiring the interaction with design engineers, will be necessary.

Future space-based gravitational-wave detectors will detect gravitational waves with high sensitivity in the millihertz frequency band, providing more opportunities to test theories of gravity than ground-based detectors. The study of quasinormal modes (QNMs) and their application in gravity theory testing have been an important aspect in the field of gravitational physics. In this study, we investigate the capability of future space-based gravitational wave detectors, such as LISA, TaiJi, and TianQin, to constrain the dimensionless deviating parameter for Einstein-dilaton-Gauss-Bonnet (EdGB) gravity with ringdown signals from the merger of binary black holes. The ringdown signal is modeled by the two strongest QNMs in EdGB gravity. Considering time-delay interferometry, we calculate the signal-to-noise ratio of different space-based detectors for ringdown signals to analyze their capabilities. The Fisher information matrix is employed to analyze the accuracy of parameter estimation, with particular focus on the dimensionless deviating parameter for EdGB gravity. The impact of the parameters of gravitational wave sources on the estimation accuracy of the dimensionless deviating parameter is also studied. We find that the constraint ability of EdGB gravity is limited because the uncertainty of the dimensionless deviating parameter increases with a decrease in the dimensionless deviating parameter. LISA and TaiJi offer more advantages in constraining the dimensionless deviating parameter to a more accurate level for massive black holes, whereas TianQin is more suited to less massive black holes. The Bayesian inference method is used to perform parameter estimation on simulated data, which verifies the reliability of the conclusion.