Recognition of a Sucker Rod's Defect with SVM

In order to improve the recognition rate of a sucker rod's defect and reduce the rapture possibility of the rod, the mixed characters include of wavelet packet energy character and the peak value in the time-domain were used to input a recognition network, and support vector machines (SVM) were used. Tested results with lots of data acquired in laboratory showed that SVM based on the mixed characters can enhance recognition rate of the sucker rod's defect.