Method for Quickly Identifying Mine Water Inrush Using Convolutional Neural Network in Coal Mine Safety Mining

In the early stage of water inrush prevention, accurate and rapid identification of water source is required to play an early warning role in water inrush prevention and control. The AF structure is a multi-layer network model between the shallow layer and the deep layer, which can reduce the original spectral data to 2 dimensions. In order to make the dimensionality reduction model sparse, a convolutional neural network layer is added to the traditional AF algorithm. First, the unsupervised learning algorithm is used to reduce the dimension of the original spectral data, so as to reduce the influence of redundant information in the spectral data on clustering. The identification of coal mine water source type runs through the early prediction and later treatment of water inrush prevention and control. Secondly, a mine inrush water source identification model of support vector machine and convolutional neural network is constructed. On this basis, an improved frog jump optimization algorithm for mine inrush water source identification is proposed to solve the local optimal solution problem caused by the randomness of initial weight setting of convolutional neural network. Compared with convolution and neural network, the recognition rate of the optimized leapfrog optimization algorithm is improved. Finally, the model is optimized from the aspects of performance fluctuation, function singleness and constraint of training mode, and combined with the demand of water source identification. The optimized model has the characteristics of anti-interference, functional expansibility and online identification, etc., and its effectiveness is verified by standard data set, which is extended to the water source spectral data, so as to assist the prevention and control of coal mine water in burst disaster. According to the experiment, the dimensionality reduction model with the addition of the convolutional neural network layer has a faster convergence rate. The recognition rate of spectral data based on DBN method is 91.07% s and recognition rate of 99.02%.