Reliability Assessment of Machining Accuracy on Support Vector Machine.

In order to establish the reliability model of limited samples, a kind of support vector machine (SVM) tools LSSVM.M is used to process reliability assessment data. The choice of parameters is the key factor to affect the forecasting accuracy. To improve the accuracy, a method of advanced parameters choice is proposed, which consists of two steps. First step is to optimize the scope of parameters, gradually narrowing the initial large scope. The next step is to find out the optimal parameters. Finally, the reliability model of NC machining accuracy was established, and the prediction result in NC machining accuracy case proves that this method is valid and effective. [ABSTRACT FROM AUTHOR]