reliability prediction
1.The reliability design method for the electrical system of a car is studied by the calculating the reliability indexes of the electrical system,modeling for reliability of the designed circuit,verifying the reliability indices with the rapid prediction method,selecting componeuts and allocating reliability,and verifying the reliability of the circuit through reliability prediction.
2.On the base of reliability prediction,a math model is built for the fuzzy distribution to the dif- ference between the values of prediction and objective. A new method is given to gun reliability distribu- tion by combining values of the difference and prediction. An auto cannon is exampled for illustration analysis.
3.Thereout, we introduce SVM theory into the field of software reliability early prediction, and advance the software reliability early prediction model based on SVM with the help of many mathematic tools such as clustering analysis, Primary Components Analyse (PCA), discriminant of strange samples, data scaleing, cross validation, non-enlighten Depth-First Search (DFS) and so on.
6.The detail fatigue rating (DFR) method, based on probability fatigue, is one of fatigue reliability life design and prediction methods for civil aircraft structure, and has been widespreadly developed both home and abroad.
7.Presents the neural network based prediction of software faults in integrated navigation system including the prediction of number of faults by a single layer of network only to reduce training time, and a new static reliability model suitable for prediction of quality and quality fit depending upon complexity of software, and the comparison with the traditional regression technique which shows that the prediction with a single hidden layer neural network achieves better quality than that with the traditional statistic model, and concludes that the neural network technique has better quality fit and prediction capability when it is applied to the data for a large-scale integrated navigation system and it provides a good managemant tool for software engineering.
10.In view of the parameters characteristics and their relation with the contact reliability, abstractly base on the nonlinear smoothing prediction model (time series analysis) and the two-level fuzzy synthetical evaluation model (fuzzy synthetical evaluation method), advance the synthetical evaluation and prediction method of the contact reliability of contacts (contact system state), quantificationally study the contact reliability in primary.
