Identify Thin Reservoir Thicker Than One Meter by Integration of Seismic and Logging Data

Liqin Gan,Jianbo Chen,Yawei Hou,Yue Xie,Hongfu Shi
DOI: https://doi.org/10.2523/iptc-23360-ea
2024-01-01
Abstract:Abstract As the oilfield development progresses into the middle and later stage, thick sand reservoir with good quality has reached high water-cut. More attention has been paid in thin reservoir development but its characterization is very difficult. At present, the lower limit of reservoir thickness detected by seismic is five meter. This paper proposed a new method to identify thin sand thicker than one meter by integration of seismic and logging data. Seismic data has a high lateral resolution which can reflect lateral changes of reservoir, while logging data has a highed to integrate logging and seismic data to predict thin reservoir thicker than one meter vertical resolution which can identify reservoirs thicker than one meter. Therefore, it is propos. The first step is to screen out natural gamma ray (GR) which can differentiate sand and shale well then standardize, and transform wavelet to obtain low and middle frequency information which can reflect lithology. Secondly, well/seismic calibration, synthetic, and singular value decomposition were conducted to realize dynamic clustering of seismic waveforms. Then the common characteristics of seismic waveform and GR of the same layer and depositional microfacies were analyzed, and the database for microfacies, waveform and GR at well location was established. Finally, the initial model was obtained by simulation of the original 3D seismic volume track by track, comparison of the sample group with highest correlation, and weighted average of correlation of GR. Then the final inversion result was achieved through the iterative inversion under the guidance of Bayesian theory. This method relies on high quality of GR data and has a good characterization effect on shallow water delta sedimentation in gentle slope areas. Taking P oilfield as an example, seismic inversion by above process showed that sand reservoir thicker than one meter can be detected. Five verified wells showed that the coincidence rate of reservoir prediction can reach up to 87%.
What problem does this paper attempt to address?