基于均生函数模型的冲击矿压电磁辐射预测研究及其应用

Research on Prediction of Rock or Coal Outburst by Means of EME Method Based on Mean Generating Function Model

通过实验研究煤岩变形破裂过程电磁辐射信号的变化规律,表明非接触电磁辐射法能动态预测冲击矿压等煤岩动力灾害现象;在非接触电磁辐射法动态预测冲击矿压的实验研究基础上,提出一种新的动态时间序列预测模型——均生函数模型,利用电磁辐射监测仪测定的现场工作面电磁辐射信号时间数据序列,通过逐步回归筛选时间序列构造一个均生函数方程;利用该方程预测预报未来电磁辐射信号的发展,并与现场测定的实际值进行对比分析,以此来验证该预测模型的正确性。误差分析和实践结果表明:均生函数模型的预测值与实际测定值的相对误差最大为6.71%左右,且距平趋势正确率均达到了60%,证明该模型与电磁辐射预测方法的有机结合能有效地预测冲击矿压以及提高预测的准确性,为冲击矿压电磁辐射预测技术的研究提供了一种新的研究思路和方法。

First, the changing law of electromagnetic emission （EME） signal is studied experimentally during the course of fracture of coal or rock by means of electromagnetic emission instrument. The experimental results show that the non-contacting EME method can efficiently predict such dynamical disasters as rock or coal outburst. Then, on the basis of the above experimental research, a new dynamic data model, mean generating function prediction model, has been p by use of an observed data series of EME signals in this paper. By use of the stepwise regression method, a mean generating function equation which is retrieved from the time data series of in-situ EME signal is established. Finally, the above forecasting equation is used to predict the EME data in the future, and the prediction fitting value is compared with the practical data. Some computing cases are given which show that the forecasting accuracy of the mean generation function model is satisfactory, the relative difference between the predicting value and practical measured value is under 6.71 percent, and the mean-difference trend value is about 60 percent. The organic combination of EME method and mean generation function model can predict the rock or coal out-burst efficiently. In this work, a new research idea and method is provided for prediction of rock or coal outburst based on EME prediction method.