Fault Diagnosis Based on Wavelet Entropy Feature Extraction and Information Fusion

It is important to reduce keeping costs and hold up unscheduled downtimes for machinery. So knowledge of what, where and how faults occur is very important. Fault detection and diagnosis are necessary for implementing CBM (condition base model) Best classifier systems which are considered as one of the most significant advances in pattern classification in recent years. We exposure new algorithm in this paper, this new algorithm have 3 steps. In the First step used wavelet Entropy for make wavelet tree with coefficient in each node. In second step using wavelet tree fused data with maximum coefficient in wavelet tree and in step three with output of fusion function we classification this fusion data by kernel method. This algorithm have best time study because the time of search algorithms is, D is depth of wavelet tree. Our proposed fusion strategies take into account that a Wavelet-Entropy by finding the optimal kernel size with maximal margin. Then a kernel Machine classifier is trained.