Evaluation of nuclear equipment technical condition based on support vector machine

It is difficult to evaluate the technical condition for complicated structure, lack of samples and condition data. In order to solve the problem, a method based on support vector machine (SVM) which had its own advantages of solving the classification and evaluation in the case of limited examples is given. Take the canned motor pump (CMP) for example, the indices´ grade model and code coding rules are established, and the technical condition is evaluated by SVM with different kernel functions. The results show that SVM especially with RBF kernel function can get faster calculating speed, high generalization capability and more exact result.