A Dynamic Coal-rock Interface Recognition Model Based on Cutting Acoustic Emission and Fuzzy D-S Theory

This paper presents a new dynamic recognition model to identify the coal-rock interface, which overcame the problems of low identification accuracy and hysteresis quality. The proposed recognition model designed on the basis of cutting acoustic emission and fuzzy D-S theory is able to confirm the ratio between coal and rock. For this purpose, seven coal-rock specimens were poured with different ratios. Thus, the acoustic emission feature signals were tested in the process of cutting the given specimens. Furthermore, by using the wavelet packet decomposition and energy reconstruction, the energy distributions in different frequency bands were obtained to extract the regular frequency-band energy along with the change of coal cutting ratio. In accordance with the extracted frequency-band energy, establishing the coal-rock interface recognition model based on minimum fuzzy entropy as well as combining with the D-S theory. Finally, a cutting experimentation of a random coal-rock interface was carried out to verify the accuracy of the identification model. The proposed identification model provides a technical prerequisite for automatic and intelligent height control.