Full waveform microseismic inversion using differential evolution algorithm


An accurate and fast estimation of microseismic activities from passive microseismic data is a crucial issue to many oil and gas applications. Traditional methods are mainly based on manual picking of the arrival times which require a relative high signal-to-noise ratio (SNR) to produce reliable results. When the sensor array is deployed on the surface, the microseismic events have low magnitude and might be buried in strong ambient noise. A compressive sensing scheme has been introduced to implement seismic source parameters estimation, including location (i.e., hypocenter) and moment tensor (MT). Although this scheme is efficient and accurate, it entails computing the whole dictionary composed of Green's functions in advance, which brings a high computational overhead and prohibitive storage burden. In this work, we propose the differential evolution (DE) algorithm for solving the inverse problem on the fly which avoids generating and storing the whole dictionary.

Orlando, FL