Compared with ordinary rolling bearing,flexible thinwall bearing has background impact load and makes fault feature extraction difficult.To solve the problem,a fault feature extraction method based on MKurt-MOMEDA and Teager energy operator was proposed for flexible thinwall bearing.The kurtosis spectrum (MKurt) was used to analyze the original fault signal to determine the flexible thin wall bearing failure cycle,then it was processed by using the optimal minimum entropy solution through multipoint convolution (MOMEDA),and then it was enhanced through Teager energy operator.Finally,the fault characteristic frequency of the flexible thin wall bearing outer ring and inner ring was extracted.The fault feature extraction method based on MKurt-MOMEDA and Teager energy operator was compared with a single MOMEDA algorithm.It is proved that using the method,the fault characteristic frequency amplitude is obviously enhanced.It provides reference for fault feature extraction of flexible thinwalled bearings.