Fault Diagnosis of Engine Based on Supervision of Data-Driven

Several kinds of information generated during the operation process of the engine, generally speaking, there is no one-to-one correspondence between the characteristic parameters and status, whereas, there are often existing many types of faults. After analyze the problem of uncertainty and other issues which the fault diagnosis of engine is faced, in this paper, a new approach for fault diagnosis of engine based on supervision of data-driven is proposed. This algorithm begin with the given classification data, using the representative points on behalf of class mean values, using the weighted distances in place of Euclidean distances. Then employing the method to identify 8 kinds of common fault states for engine, the experiment results shows that the method based on optimal representative points clustering is an effective way to diagnosis the fault for engine.