Abstract. With the development of modern technology, the diagnosis technology for mechanical equipment has attached more attentions. This paper mainly presents the fault diagnosing process for rolling bearing by using the hybrid GA-BP algorithm, which combines genetic algorithm (GA) with error back-propagation algorithm (BP). The weights and thresholds of BP neural network are optimized by using genetic algorithm firstly and then assigned to the BP neural network. With the using of proposed approach, the diagnosis performance of BP neural network can be improved. In this paper, the optimal nonlinear mapping relationship between fault types and fault symptoms of rolling bearings is obtained and then realize failure recognition of rolling bearings though the hybrid GA-BP algorithm. Finally, the results of the test experiment verifies the efficiency and accuracy of the proposed GA-BP algorithm.