Research on Coal-Rock Recognition Method Based on Signal Detection and Algorithm Analysis
Aiming at the problem of low accuracy of coal-rock recognition with different hardness,a coal-rock recognition method was proposed based on coal-rock cutting test combined with Kalman filter algorithm and random forest algorithm.The cutting device was used to cut one coal seam and five coal-rock combinations with different hardness in the similar simulation test.The three-phase current characteristic signals of the three-phase motor under six cutting conditions and the infrared thermal imaging temperature characteristic signals of the contact surface between the simulated cutting drum and the coal-rock mass were collected.The results show that the peak values of three-phase current and infrared thermal imaging temperature increase with the increase of hardness of coal-rock mass.When cutting the same kind of coal-rock mass,the three-phase current and infrared thermal imaging temperature will increase with the increase of the contact area between the simulated cutting drum and the rock layer.Kalman filter algorithm was used to process the original samples,and then the original samples and the samples processed by the Kalman filter algorithm were classified and compared by the random forest algorithm.After processing by the two algorithms,the number of correctly predicted samples increase from 1238 to 1430,and the accuracy of coal-rock recognition increases from 84.17％to 99.38％.The research results can provide reference for intelligent precision coal cutting of shearer.
万方数据
维普
