科研之友
微信
新浪微博
Facebook
分享链接
摘要
Improvements in seismic data acquisition and processing have made seismic technology a viable source of information for locating hydrocarbon deposits and also for describing the spatial variability of reservoir parameters. While three-dimensional seismic technology is already a well-established means of locating hydrocarbon deposits, the 4D (or time-lapse) seismic is gradually becoming a source of reservoir parameter description. In particular, time-lapse seismic is increasingly becoming a useful source of information about fluid migration and pressure changes in the reservoir. In recent years, time-lapse seismic data or saturation maps derived from such data have been used to estimate spatial distribution of reservoir parameters through history matching. However, successful application of the method to accurate description of reservoir parameter variability remains a challenge. Some major challenges in the application of time-lapse seismic data in reservoir model history match are the poor resolution of the seismic data, uncertainty associated with the maps of changes in saturation derived from the seismic data, and the massiveness of the data or the associated saturation maps. Repeat seismic data are blurred or low-resolution maps of trends in reservoir responses to production and injection activities. These trends are influenced by reservoir property distribution and when used judiciously, constitute a good source of information that can help in reducing the uncertainty associated with reservoir parameter estimates. The maps of changes in reservoir saturation or pressure obtained from repeat seismic data are themselves fraught with uncertainties and care must be taken when using these maps to estimate reservoir parameters. Furthermore, obtaining the most useful information from a voluminous seismic data set is challenging and is still an active area of research. In this work, we present the use of the wavelet transform to integrate maps of changes in reservoir saturation derived from time-lapse seismic data into reservoir model history matching. The work involves transforming the saturation-change map into a wavelet space and then history matching some carefully selected wavelets to obtain estimates of reservoir parameters. Three sample applications based on synthetic data are used to show the suitability of the approach. Comparison is made to conventional approach in which the saturation-change map is not transformed into wavelets before matching.
- 出版日期2014-10
- 单位中国石油大学（北京）
