The existing convex hull based support vector machine (SVM) algorithm has too long time in processing large-scale original data generated by mechanical equipment. To solve this problem, by combining Alpha Shapes algorithm and fast convex hull algorithm, an SVM combined with improved fast convex hull algorithm was proposed for fault diagnosis.In the fusion algorithm,the improved simplified Alpha Shapes algorithm was used to extract the boundary data points of the point set as the object of the improved fast convex hull algorithm to reduce the recursive workload of the convex hull algorithm. The experimental results show that the algorithm extracts only 026% of the data points of the data set on average, and the calculation time is reduced accordingly. Finally, the experimental results also show that the performance of the algorithm is better than a single SVM algorithm.