Abstract: In order to improve prediction accuracy of coal seam water infusion effect by using BP neural network, genetic algorithm was used to optimize weight value and threshold value of BP neural network. Genetic algorithm-BP neural network model was built and used to predict wetting radius of coal seam water infusion. The Matlab simulation result shows that the genetic algorithm-BP neural network model has more accurate prediction result than BP neural network model, average relative error of the genetic algorithm-BP neural network model is reduced by 40.29%, training steps are reduced by 1 665 steps, convergence speed is fast and stability is good. Key words: