The Fault Diagnosis of Elevator Based on the Autoregressive Model And the Support Vector Machine

The paper proposed the fault diagnosis of elevator based on the autoregressive (AR) model and the support vector machine. At first, build the AR model with the processing signals. The AR coefficient was used as the input of the support vector machine, the normal condition and the fault condition were used as output. By studying and predicting of the support vector machine (SVM) can reaching automatic identification. This method has high accuracy of diagnosis while resolved the problem of lacking samples. With the fast city construction, elevator as vertical vehicle was applied more and more widely, which has complicated structure and required high reliability At present, many accidents occur during the elevator working, such as people get trapped, elevator slipped, and so on. So elevator operation reliability testing needed to be improved. Improving the operation reliability, on the one hand, modify design and improving the installing quality; the other hand, rely on the advanced technique of fault diagnosis([1]).