Application of machine learning to microseismic event detection in distributed acoustic sensing data
Application of machine learning to microseismic event detection in distributed acoustic sensing data
复制标题
DOI：
10.1190/geo2019-0774.1
复制
发表时间：
2020-09
期刊：
影响因子：
3.3
通讯作者：
A. Stork;A. Baird;S. Horne;G. Naldrett;S. Lapins;J. Kendall;J. Wookey;J. Verdon;A. Clarke;Anna Williams
中科院分区：
文献类型：
--
作者：
A. Stork;A. Baird;S. Horne;G. Naldrett;S. Lapins;J. Kendall;J. Wookey;J. Verdon;A. Clarke;Anna Williams
这项研究首次展示了卷积神经网络（CNN）的可转移性，该网络被训练用于检测一个光纤分布式声学传感（DAS）数据集到其他数据集中的微震事件。DAS越来越多地用于工业环境中的微震监测，这些系统提供的密集空间和时间采样产生大量数据（对于2公里长的电缆，以2000 Hz采样，空间采样为1 m，每天约650 GB），需要新的处理技术进行近实时微震分析。我们已经训练了被称为YOLOv 3的CNN，这是一种对象检测算法，使用叠加了真实的噪声的合成生成波形来检测微震事件。CNN网络的性能与使用过滤和幅度阈值（短期平均/长期平均）检测技术检测到的事件数量进行比较。在提取真实的噪声的数据集中，该网络能够检测到>80%的通过人工检查识别的事件，并且比通过标准频率-波数滤波技术检测到的事件多14%。错误检测率约为2%，即每20秒发生一次事件。在其他数据集中，通过监测网络以前看不到的几何形状和条件，人工检查识别的事件中有>50%被CNN检测到。
This study presents the first demonstration of the transferability of a convolutional neural network (CNN) trained to detect microseismic events in one fiber-optic distributed acoustic sensing (DAS) data set to other data sets. DAS increasingly is being used for microseismic monitoring in industrial settings, and the dense spatial and temporal sampling provided by these systems produces large data volumes (approximately 650 GB/day for a 2 km long cable sampling at 2000 Hz with a spatial sampling of 1 m), requiring new processing techniques for near-real-time microseismic analysis. We have trained the CNN known as YOLOv3, an object detection algorithm, to detect microseismic events using synthetically generated waveforms with real noise superimposed. The performance of the CNN network is compared to the number of events detected using filtering and amplitude threshold (short-term average/long-term average) detection techniques. In the data set from which the real noise is taken, the network is able to detect >80% of the events identified by manual inspection and 14% more than detected by standard frequency-wavenumber filtering techniques. The false detection rate is approximately 2% or one event every 20 s. In other data sets, with monitoring geometries and conditions previously unseen by the network, >50% of events identified by manual inspection are detected by the CNN.