中文摘要：

通过对合成微震数据和实测微震记录的处理实验,对自动检测有效微震信号的STA/LTA（短时窗平均/长时窗平均）算法及其改进的加权系数法、多窗口算法和修正的能量比算法进行了分析和对比,给出了时窗长度、触发阈值和特征函数对算法性能的影响特征及其选取规律.与原STA/LTA算法相比,加权系数法,降低了微震事件的漏判率;多窗口算法和修正的能量比法提高了对低信噪比微震记录检测的正确率及微震到时的拾取精度.

英文摘要：

Through tests on synthetic and observed microseismic records,the STA/LTA（short term averaging/long term averaging）algorithm for automatic detection of microseismic signals and its improved algorithms including weighting factor method, multi-window algorithm, and modified energy ratio algorithm are analyzed and compared,and the performance of the algorithms and selection rules for the window length,the trigger threshold and the characteristic function are then obtained.Compared with the STA/LTA algorithm,the weighting factor method can reduce the error rate of microseismic detection,and the multi-window algorithm and the modified energy ratio algorithm can improve the accuracy of microseismic events detection for low signal-to-noise ratio microseismic data.