Aiming at the problem that BP neural network have a slow convergence speed and is easy to fall into local minimum when applied to fault diagnosis, a fault diagnosis method of hydraulic drilling rig based on Genetic algorithm optimized BP neural network is proposed. Using selection, crossover and mutation to optimize the weights and thresholds of BP neural network, and improve the convergence speed of network training and the optimized BP neural network is applied in the fault diagnosis of hydraulic drilling rig. Extract appropriate fault feature from the collected parameters of hydraulic drilling rig working condition, then establish sample set by data normalization, using the training sample set to train the network, and finally, make fault diagnosis according to the training results. The simulation reveals that the BP neural network optimized by genetic algorithm has a small number of iterations, and a fast convergence speed. It can classify the testing samples effectively, and it has a high accuracy of fault diagnosis.