ARRIVAL TIME PICKING OF MICROSEISMIC DATA BY USING SPE ALGORITHM

摘要

The accuracy of the arrival time picking has deep influence on hypocenter location which is the core part of microseismic exploration, so arrival time picking blays a significant role in microseismic date brocessin. However, at this stae, the arrival nicking of microseismic siumls with low siuml-to-noise ratios (SNR) are problematic because these valid signals are usually obscured by random noise, so it is difficult to obtain the arrival time of microseismic signal accurately by the conventional methods, especially the horizontal components in microsiesmic signals. In order to solve this technical issue, the valid signals in microseismic data is highlighted by the multi-scale and multi-direction features of Shearlet transform, and because of the vibration track's difference between the valid signals and noise, the polarization analysis is applied to process the Shearlet coefficients. Also, through the introduction of entropy and the proposal of weight factor, three-dimensional (3D) weighted entropy algorithm is constructed, so as to achieve the purpose of processing the three components together. Several synthetic and field data examples with different kinds of noise demonstrate the effectiveness and robustness of the Shearlet transform-polarization analysis-3D weighted entropy (SPE) algorithm in arrival time picking of microseismic data.