Identification of coal-rock interface based on principal component analysis and BP neural network

In view of problem of slow identification speed and bad real-time performance of current identification method of coal-rock interface because of extracting much time domain signals, an identification method based on principal component analysis and BP neural network was proposed. According to the method, time-domain signals of cutting torque of shear drum were selected at first, and then PCA method was used to compress the signals. At last, these final signals were input into BP network to identify coal-rock interface. The simulation result shows that the method can not only meet with recognition rate, but also increase identification speed, which establishes foundation for improving response of drum lifting.