矿井涌水量时间序列的长程相关性分析及分维数估算

Analysis of long-range correlation and fractal dimension estimation of time series of mining water inflow

根据矿井涌水量实测数据,探讨了矿井涌水量时间序列的两个重要特性,即:长程相关性和统计自相似特性。利用小波分析和分形理论在多尺度分析和自相似本质上的一致性,把小波分析和分形理论引入到矿井涌水量时间序列的分析中,得到矿井涌水量时间序列小波变换系数,在此基础上,提出了矿井涌水量时间序列分形维数的小波计算方法,并对巷道涌水量和回采工作面涌水量时间序列的分维值进行了计算,验证了小波分形维数估计法在提取矿井涌水量中所具有的分形特征信息是稳定的和可靠的。

Based on field measurements of mining inflow,this paper examines two important characteristics of the mining inflow,that is,the long-range correlation and the statistical self-similarity. By using the consistency of multi-scale analysis and self-similarity of the wavelet analysis and fractal theory,the wavelet analysis and fractal theory are used to analyze self-similar characteristics for the time series of mining inflow. Wavelet transform coefficients of the time series of the mining inflow are obtained with the orthogonal wavelet transform. The calculation method with wavelet and its steps to calculating fractal dimensions of the time series of the mining inflow are presented. The fractal dimension values of the time series of the water inflow in roadways and working face of a mining section are calculated. The research results demonstrate that the method is stable and reliable for obtaining fractal characteristics of mining inflow.