Abstract: A fuzzy recognition method based on weighted fuzzy relative entropy is proposed to recognize motor rotor fault,and it introduces weighting idea into fuzzy relative entropy to identify the fault severity level of motor rotor.With the introduction of the weighting method,the proportion of fuzzy relative entropy of the symbolic regions with abundant information would be increased among the fuzzy relative entropies of the whole region,and with the fault information included in symbolic regions with abundant information,the motor fault can be recognized more sufficiently and reasonably. Simulation results of fault diagnosis on broken rotor in induction motor show that the proposed method can effectively realize quantitative analysis of motor faults and can accurately recognize the motor fault level.In addition,the algorithm robustness is improved,and the reliability and the accuracy of motor fault classification are enhanced.