微震与爆破事件统计识别方法及工程应用

 采用人工识别方法，依托微震监测系统，建立矿山爆破与微震事件样本数据库。统计分析数据库内各事件地震力矩、事件总能量、事件P波能量与S波能量比、事件的静压力降、事件的发生时间、传感器触发数量和拐角频率等震源参数特征；对比分析首次峰值到时、首次峰值幅值、最大峰值到时及最大峰值幅值的概率密度分布特征；通过FFT变换，统计分析2类事件信号的主频分布规律。依据各参数的概率密度分布及其识别效果，结合特征参数获取的难易程度，最终选取事件的地震力矩对数，事件的能量对数，事件的传感器触发数量，首次峰值幅值对数及最大峰值到时对数和信号的主频为特征参数，建立矿山微震与爆破事件自动识别的统计学模型。该模型对样本的回检结果显示，50组建模样本的准确率为100%，50组测试样本的准确率为94%。将该模型应用于采场大块矿石的二次破碎事件识别中，识别结果与实际相符，解决了单纯依靠信号特征识别导致该类事件极易与微震事件混淆的问题。该方法误判率低，特征参数较易获取，是矿山微震与爆破事件识别的一种有效方法，可在实际工程中推广应用。

A statistical method to identify blasts and microseismic events and its engineering application

A database of blasts and microseismic events was established with the manual recognition methods based on the microseismic monitoring at Kaiyang phosphate mine. The source parameters including the seismic moment，the seismic energy，the P and S wave energy ratios，the event occurrence time，the static stress drop，the sensor triggers and the corner frequency were analyzed statistically. The probability density distributions of the first peak arrival time，the first peak amplitude，the maximum peak arrival time and the maximum peak amplitude were compared and analyzed. The frequency distributions of two kinds of event signals were statistically analyzed with the FFT transform. The logarithm of seismic moment，the seismic energy，the event occurrence time，the first peak arrival time，the maximum peak amplitude，the numbers of triggered sensors and the dominant frequency were finally selected as the characteristic parameters based on the probability density distribution of each parameter，the performance of recognition and the difficulty of acquisition. A mathematical model of automatic recognition was established on the application of Fisher discriminant analysis. Results show that the accuracy of training samples is 100% and that the accuracy of testing samples is 94%. The model was applied to the second time blasting crushing of bulky rock of the stope and the recognition results were consistent with the actual ones.