基于多重分形理论的工作面煤与瓦斯突出预测  

Coal and gas outburst prediction in working face based on multi-fractal theory

作　　者：邓存宝[1] 凡永鹏 

机构地区：[1]辽宁工程技术大学安全科学与工程学院,辽宁阜新123000 [2]辽宁工程技术大学矿业学院,辽宁阜新123000

出　　处：《辽宁工程技术大学学报（自然科学版）》2017年第9期903-908,共6页Journal of Liaoning Technical University (Natural Science)

基　　金：辽宁省教育厅创新团队项目(LT2012009);国家自然科学基金(51174108)

摘　　要：为对工作面煤与瓦斯突出进行高效准确的预测,采用多重分形理论分析的方法,对瓦斯涌出时间序列进行研究.绘制了分配函数和划分尺度之间的双对数曲线,计算了广义分形维数D_q、奇异指数α、多重分形谱f(α),分析了k_0、?α、?f(α)在不同时间段的变化规律,从不同角度反应瓦斯涌出量的非均匀质特征.研究结果表明:无论是q>0时还是q<0时所有ε区间上的分配函数和划分尺度之间的双对数曲线都是直线,瓦斯涌出量时间序列具有多重分形特征,在较大测度内满足标度不变性.多重分形参数k0、?α、|?f(α)|在一般情况会不断小幅度的变化但都趋于稳定,在快接近煤与瓦斯突出发生时间点时显著大于正常水平.它们之间有很好的相关性,可以共同对煤与瓦斯突出进行预测.In order to accurately predict the coal and gas outburst in working face, this paper studied the time series of gas emission quantity by utilizing the method of multi-fractal. The study drew the double logarithmic curves between the distribution function and classification scale, calculated the multi-fractal parameters of generalized fractal dimension Dq, singular index a, multi-fractal spectrumfla） and analyzed the variation law of velocity at different time. The results show that the double logarithmic curves between the distribution function and classification scale are straight lines whether q〉0 or q〈0. It shows the multi-fractal characteristics of gas emission quantity time series. In general, changes of the multi-fractal parameters are small but tend to be stable. They are significantly higher than the normal level when the coal and gas outburst occurs in the near time. There is a good correlation between the multi fractal parameters, which can predict the outburst of coal and gas together.

关 键 词：煤与瓦斯突出 瓦斯涌出量 瓦斯预测 多重分形 分形维数 

分 类 号：TD713[矿业工程—矿井通风与安全]