The Reliability Evaluation Methods Of The Products Of Electrical Engineering And The Applications Of The Bayes Theory

The technology of the reliability theory and its application is a new technology. It has important effect to ensure and improve the quality of the products. In the theory research and application work of the reliability of the products of electrical engineering, we wish to find the accurate calculating formula about the reliability characters of the products of electrical engineering easily. But the reliability characters of the products of electrical engineering are determined by their failure distribution. We can not calculate the reliability characters of the products unless we know their failure distribution. In the actual works, the failure distribution of the products of electrical engineering can not be determined easily. So we can hardly calculate the reliability characters of the products of electrical engineering accurately. Therefore, it is particularly significant to find the easy and accurate reliability evaluation methods in the actual works. In this paper, the evaluation methods of the reliability characters of the products of electrical engineering are studied.For the first time, a new approach is developed to estimate the Weibull distribution' s parameters by using the Median rank and the least梥quares method. The estimation of the confidence limits of the mean time to failure is given. A actual test data is computed by using the proposed method.A new approach is developed to estimate the mixed梂eibull distribution ?s parameters using the Bayes' theory. At first, the population sample data are split into subpopulation data sets over the whole test duration by using the posterior belonging probability of each observation to each subpopulation. then, with the concept of accumulated failure probability, the proposedapproach combines the Least梥quares method with Bayes' theorem, takes advantage of the parameter estimation for single Weibull distribution to each derived subgroup data set, and estimates the parameters of each subpopulation. The estimates given by this paper also satisfy the Maximum Likelihood Equation. The mean time to failure and the reliability estimation of the mixed population are given. The algorithm and the computing flow chart of the proposed method are given. A actual test data is computed by using the proposed method.For the first time the Tikhonov' s regularizing method is used to solve the failure rate of the products of electrical engineering when their failure distribution is unknown. The algorithm of the proposed method is given. A numerical example is given to use the proposed method. It turns out that the proposed method yields more accurate result.In this paper, a form of prior distribution about the fail梡robability pl in time f, of zero梖ailure date(/,,?)is developed reasonably. Basing on the prior distribution, we obtain theBayesian estimations about/*,in timer,. Finally, Calculation is performed regarding to practical problem. It turns out that the proposed method accords with the engineering priors.