Prediction model of water conducted fractured zone height based on the PSO-RBF neural network

The selection accuracy of the water conducted fractured zone height directly affects the safety of mining under the water body.To improve the accuracy for the water conducted fractured zone height，in the，this paper selects 6 main factors are selected as the input layer neurons based on the comprehensive analysis of the main influence factors of the water conducted fractured zone height.And the POS algorithm and radial basis function neural network are organically combined，the PSO-RBF neural network water conducted fractured zone height prediction model is built.The reliability of the model is verified by the training and testing of 25 sets of measured data.The results show that，compared with the measured results，the maximum relative difference of the predicted results is 7.43% and the minimum is 1.41%.