In order to solve the problem that the accuracy of multi-type fault classification of motor bearing is not high, a fault diagnosis method of motor bearing based on the improved Aquila optimization algorithm (IAO) is proposed, which is used to optimize the Support vector machine of motor bearing.Firstly, the basic Aquila optimization algorithm is introduced, and then Tent chaotic map and adaptive weight are introduced to improve the algorithm. Secondly,VMD is performed on the time domain signal samples of rolling bearing faults under 10 states, and the time and frequency domain features of different states are obtained. Finally, the penalty parameter (C) and kernel parameter (g) of support vector machine were optimized by the IAO algorithm, so as to construct the IAO-SVM rolling bearing fault diagnosis model. The final results show that the IAO-SVM model has a high accuracy of 100% in fault diagnosis under 10 states of motor rolling bearings.