An improved decision-making 1-SVM algorithm with abnormal samples and its application

Aiming at normal samples abundance and abnormal samples deficiency,as adding abnormal samples could improve classification ability and classification accuracy,an improved decision-making 1-SVM algorithm with abnormal samples was put forward and applied in abnormal condition detection of mechanical equipments. On the one hand,the 1-SVM model was trained with two kinds of samples to improve the description ability of the 1-SVM algorithm for abnormal samples. On the other hand,the decision boundary was adjusted to improve the classification accuracy of the1-SVM algorithm. The improved 1-SVM algorithm was applied in fault detection of diesel engin valve train. The experimental results showed that recognition rate of the improved algorithm for normal class and fault class samples is higher than that of the standard 1-SVM algorithm and the 1-SVM algorithm only with abnormal samples.