关键词: 综采工作面, 煤岩截割状态, 小波包, 频带方差, LVQ神经网络

Abstract: In order to recognize the cutting state of shearer accurately, a pattern recognition method based on wavelet packet decomposition and learning vector quantization (LVQ) neural network is proposed. The vibration signal is decomposed by wavelet packet to realize the preprocessing of vibration signal and obtain several sub-bands. On this basis, the variance of each frequency band is calculated and used as the eigenvector. Then, the calculated frequency band variance is taken as the eigenvector and input to LVQ neural network to recognize the coal cutting state of shearer. The experimental results show that the method can realize the recognition of typical coal rock cutting state of shearer, and the average recognition accuracy is high, which is of great significance to realize the "unmanned" of fully mechanized working face.