Fault diagnosis of automobile main reducer based on correlation dimension

The theory of correlation dimension computation based on GP is concise, but the computation burden is heavy, and scaling region recognition automatically is hard. A method to scaling region recognition and correlation dimension computation automatically based on second derivative of correlation integral is presented. In theory, the second derivative of scaling region of correlation integral is zero, searching this continuous zero vicinity in the second derivative curve of correlation integral, which corresponding with the scaling region of correlation integral. The effectiveness of this method was verified by the analysis of Lorenz attractor. In addition, the correlation dimensions of signals in different conditions sampled in an automobile main reducer performance test bed were computed by this method. Experiment results show that correlation dimensions are separable between different main reducers, so correlation dimension can be used as a quantitative criterion for recognizing fault property and level, and the improved algorithm of correlation dimension is promising in the online product testing.