Study On The Dynamic Characteristics Of Gas Emission In Working Face

Gas emission in working face is the main source of mine gas emission. Abnormal gas emission can lead to the disaster of gas explosion happen besides the other tragic accidents. Most study of coal mine monitoring data is concentrated on the data itself at present. The study is not well combined with the actual production in mines, which can not make full use of the data. Therefore, know and master the gas emission dynamic characteristics in the process of extraction and to forecast gas emission accurately is of great theoretical and practical significance to improve mine safety.First of all, gray related degree analysis is applied to analyze the gas emission parameters such as coal seam gas content, exploitation level, thickness, dip angle, mining intensity factors etc. It is concluded that coal seam gas content and exploitation level are the main influence factors in working face of Guhanshan Coal Mine. The exploitation levels and the corresponding gas emission data are fitted by use of the mathematical function model. The relationship between them accords with power function law. The obtained regularity can forecast the gas emission quantity to some degreeSecondly, analyzing the collected ventilation, extraction and monitoring data on production field by MATLAB. It is concluded that the specific dynamic characteristics of gas emission in working face. The characteristics includes the gas emission change along with the recovery process is M-type; the gas emission and accumulated footage appear exponential function; the coal mining shifts gas emission is more than preparing mining shifts during normal stoping, but the latter with great fluctuation; gas emission has high volatility during drilling and has big quantity with small fluctuation during preparing prediction and discharge hole; gas emission will suddenly grew during moving the cannon; the gas emission of every shift will appear peak characteristics and present some certain regularities.Then, the gas concentration samples can be divided into three categories based on the dynamic characteristics: two consecutive days, coal mining shifts and preparing mining shifts. It is concluded that the power spectrum indexes of all samples correspond to 1 / f distributions; The C-C and G-P algorithm models are used to reconstruct the phase space of all the samples. And we get the delay time and embedding dimensions; The correlation dimensions are all scores and the largest Lyapunov indexes greater than zero of all samples, which qualitatively and quantitatively demonstrate the gas emission system of Guhanshan Coal Mine has the chaotic coupling characteristics.Finally, we obtain the Hurst indexes using R/S method and calculation the effective correlation length of all samples. The length of time is 32, 42, 33; we estimated the predicted time as long as possible with the longest some 21, 31, 27 by using the reciprocals of the maximum Lyapunov indexes. The prediction methods of weighted zero-order local-region method and adding-weight one-rank local-region method and based on the maximum Lyapunov index method are applied to predict gas concentration time series. The predicted results of absolute errors are less than 15%, 9%, 6%. The application of based on the maximum Lyapunov index method is fairly effective.