Frequency-dependent AVO inversion based on sparse constrained inversion spectral decomposition
Cong Luo, Guangtan Huang, Xiangyang Li · 2017
Frequency-dependent AVO inversion (FAVO) can extract fluid induced dispersion information and is regarded as a potential hydrocarbon identification technique. However, FAVO inversion is frequency-decomposition based, which means the inversion results can be greatly affected by the resolution and accuracy of time-frequency analysis methods. Spectral decomposition based on sparse constrained inversion (ISD) is an advantage method of time-frequency domain analysis with high resolution and high accuracy. In this abstract, a numerical test through a complex synthetic signal was designed to compare ISD with other conventional methods. The test result demonstrates that ISD method has higher time-frequency resolution than linear transform (GST) and has more precise analysis result and less computing time than bilinear transform (MP-WVD). Then, we innovatively proposed a fluid identification scheme of combining ISD with FAVO inversion, and applied this scheme to two sets of real seismic data. The application results of FAVO inversion based on ISD have higher resolution, agree better with well information and provide us more credible boundary information of gas- and oil-saturated area. All these demonstrate the advantage and reliability of this scheme in hydrocarbon detection. Presentation Date: Tuesday, September 26, 2017 Start Time: 8:55 AM Location: 370D Presentation Type: ORAL