Identification of the reservoir using seismic inversion based on particle swarm optimization method: A case study
Ravi Kant,Brijesh Kumar,S P Maurya,Nitin Verma,Ajay P Singh,G Hema,Raghav Singh,K H Singh,Piyush Sarkar
DOI: https://doi.org/10.1007/s12040-024-02436-7
2024-11-13
Journal of Earth System Science
Abstract:Accurate reservoir characterization is a crucial step for developing, managing, and optimizing hydrocarbon production. In this current study, we employ particle swarm optimization techniques (PSO) to perform an inversion of post-stack seismic data, extracting information about subsurface acoustic impedance and porosity. Conventional seismic inversion methods predominantly employ local optimization strategies, which often rely on the availability of initial models, particularly in unfamiliar geological settings. In contrast, our approach is based on global search principles, consistently striving to converge towards a global optimal solution, independent of the initial model. To validate the developed technique, we initially subjected it to synthetic data and a wedge model, followed by its application to real data from the Blackfoot field in Canada. The investigation reveals that the inverted results, both for the synthetic and real data, closely align with the observed data. Statistical analysis indicates a high correlation of 0.99 for the synthetic data. For the real data, the correlation remains strong at 0.89. Finally, the PSO-based inversion algorithm is applied across the entire seismic volume, successfully yielding high-resolution subsurface information. This inversion reveals impedance variations ranging from 6500 to 13000 m/s*g/cc, with porosity levels spanning 5–24%, within the Blackfoot region. As per the findings of the investigation, it is evident that the upper section of the subsurface mainly comprises non-solid rock materials. The examination of inverted sections has disclosed an atypical region characterized by low impedance ( 18%) within the time interval of 1040–1060 ms two-way travel time. This distinctive zone is corroborated by well-log analysis at the same depth and is categorized as a reservoir.
geosciences, multidisciplinary