Application of Support Vector Machine in Fault Diagnosis of Certain Type Engine

For the engine fault diagnosis in real problems, the number of samples available are limited, and the progress of research on the theory of the most limited to assume that the data samples, so that the network training data examples, in engineering applications has been slow, in this paper, the application of support vector machine in fault diagnosis of engine, the segmentation of the training sample set, in order to achieve the optimal analysis of the machine, the reasoning ability best. First introduced the two classification method of support vector machine and multi classification method based on two classification methods of the study, and applied to the fault diagnosis of engine, and then the simulation test for this method, and compared with the existing methods, the results show the effectiveness of the classification method, the results of the analysis also can use the tree diagram or table form, simple and intuitive; but also can save the contribution to some extent in time. Copyright © 2014 IFSA Publishing, S. L.