Comparison of neural networks techniques to predict subsurface parameters based on seismic inversion: a machine learning approach

Verma, Nitin; Maurya, S. P.; kant, Ravi; Singh, K. H.; Singh, Raghav; Singh, A. P.; Hema, G.; Srivastava, M. K.; Tiwari, Alok K.; Kushwaha, P. K.; Singh, Richa

doi:10.1007/s12145-023-01199-x

Cited by 4 publications

(2 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rationale behind choosing GA as global optimization and PS as local optimization lies in their ease of implementation and lower expertise requirements. The steps in combining global and local optimization in this study include selecting seismic and well-log data, converting depth to time, implementing genetic operators to obtain an initial population, calculating reflectivity and synthetic traces, assessing RMS error between synthetic and input seismic data, and iteratively modifying the initial population to minimize RMS error within a limited time [45][46]. Subsequently, a pattern search algorithm is employed using the output of the genetic algorithm as the initial model.…”

Section: Hybrid Optimizationsmentioning

confidence: 99%

Reservoir Characterization Using Seismic Inversion Based on Sparse Layer Reflectivity and Hybrid Genetic Algorithms: A Comparative Case Study of Blackfoot, Canada

Verma,

Kant,

Singh

et al. 2023

Int. J. Petrol. Technol.

Self Cite

View full text Add to dashboard Cite

This research paper introduces a comparative case study on reservoir characterization through seismic inversion techniques. The study specifically explores sparse layer reflectivity and a hybrid approach involving genetic algorithms and pattern search. The research assesses the effectiveness of these methodologies in delineating subsurface properties, with a particular focus on acoustic impedance. Through meticulous analysis, the paper aims to identify the strengths and limitations of each method, considering factors such as parameter estimation precision, computational efficiency, and adaptability to complex geological structures. The findings contribute valuable insights for selecting optimal seismic inversion techniques in reservoir characterization, advancing our understanding of how the integration of sparse layer reflectivity and hybrid genetic algorithms can enhance subsurface imaging accuracy and reliability. The results obtained from our inversion process significantly enhance the interpretation of seismic data by providing detailed insights into the subsurface. Both the sparse layer reflectivity (SLR) and hybrid genetic algorithm (HGA) algorithms have exhibited outstanding performance when applied to real datasets. The inverted impedance section reveals notable low acoustic impedance ranging from 8000 to 8500 m/s g/cc. This distinct zone, identified as a reservoir (sand channel), is located within the time interval of 1040–1065 ms. Our observations indicate that HGA demonstrates superior correlation results not only in the vicinity of well locations but also over a broader spatial range, suggesting its potential to provide higher-resolution outcomes compared to SLR.

show abstract

Section: Hybrid Optimizationsmentioning

confidence: 99%

Reservoir Characterization Using Seismic Inversion Based on Sparse Layer Reflectivity and Hybrid Genetic Algorithms: A Comparative Case Study of Blackfoot, Canada

Verma,

Kant,

Singh

et al. 2023

Int. J. Petrol. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Using post-stack 3D seismic amplitude data, Leite and Vidal (2011) made porosity maps that consider constraints from borehole log density and acoustic data. The impact of DL on predicting rock properties has been a subject of study in several works, comparing various algorithms (Al-Anazi & Gates, 2012;Nourani et al, 2022;Singh et al, 2024;Verma et al, 2024). Specifically, Al-Anazi and Gates (2012) compared SVR and MLP for evaluating porosity and permeability in heterogeneous sandstone reservoirs.…”

Section: Introductionmentioning

confidence: 99%

Ensemble of Neural Networks Utilizing Seismic Attributes for Rock‐Property Inversion With Uncertainty Estimation

Ntibahanana,

Jianguo,

Luemba

et al. 2024

Earth and Space Science

View full text Add to dashboard Cite

Seismic inversion holds significant importance across various domains of geoscience and engineering, including the characterization of energy resource reservoirs, the assessment of polluted sites, and CO2 storage. It is a process of estimating rock properties from seismic data that is inherently uncertain, nonlinear, non‐unique, and highly challenging. Using multiple seismic attributes increases the size of the data, requiring considerable processing resources and time. However, deep learning can accurately fit quantities of nonlinear variables, making it an excellent method for predicting spatially distributed subsurface properties. We trained some multi‐output regression neural networks to carry out porosity inversion from seismic data. We initially computed a series of seismic attributes and generated the corresponding porosity using interpreted horizons, well logs, and seismic data. Subsequently, we proposed a technique to identify the most relevant seismic attributes for porosity inversion. Because our networks work as stochastic modeling entities, we created a weight‐averaging ensemble approach to build a strong model with the highest level of accuracy. We combined realizations from baseline entities, considering their respective performance levels. Using the statistics between these realizations and the robust model, we determined the degree of uncertainty associated with the outcome. We found an R2 of 0.993 and an MAE of 0.00112 in the F3 block offshore the Netherlands, proving the method's effectiveness. The mean porosity was 0.175193, compared to 0.175626 from a reference model, and the mean uncertainty was ±0.0008998.

show abstract

Contribution to advancing aquifer geometric mapping using machine learning and deep learning techniques: a case study of the AL Haouz-Mejjate aquifer, Marrakech, Morocco

El Mezouary,

Hadri,

Kharrou

et al. 2024

Appl Water Sci

View full text Add to dashboard Cite

Groundwater resources in Morocco often face sustainability challenges due to increased exploitation and climate change. Specifically, the Al-Haouz-Mejjate groundwater in the Marrakesh region is faced with overexploitation and insufficient recharge. However, the complex subsurface geometries hamper hydrogeological modeling, characterization, and effective management. Reliably estimating aquifer substrate topography is critical for groundwater models but is challenged by limited direct measurements. This study develops nonlinear machine learning models to infer substrate depths by fusing sparse borehole logs with regional geospatial data. A Gaussian process regression approach provided robust holistic mapping, leveraging flexibility, and uncertainty quantification. Supplementary neural network architectures focus on isolating specific variable relationships, like surface elevation–substrate. Model accuracy exceeded 0.8 R-squared against validation boreholes. Spatial visualizations confirmed consistency across landscape transects. Elevation and piezometric data proved most predictive, though multivariate inputs were required for the lowest errors. The results highlight the power of statistical learning to extract meaningful patterns from disparate hydrological data. However, model opacity and the need for broader training datasets remain barriers. Overall, the work demonstrates advanced machine learning as a promising avenue for illuminating complex aquifer geometries essential for sustainability. Hybrid approaches that use both data-driven and physics-based methods can help solve long-standing problems with hydrogeological characterization.

show abstract

Comparison of neural networks techniques to predict subsurface parameters based on seismic inversion: a machine learning approach

Cited by 4 publications

References 43 publications

Reservoir Characterization Using Seismic Inversion Based on Sparse Layer Reflectivity and Hybrid Genetic Algorithms: A Comparative Case Study of Blackfoot, Canada

Reservoir Characterization Using Seismic Inversion Based on Sparse Layer Reflectivity and Hybrid Genetic Algorithms: A Comparative Case Study of Blackfoot, Canada

Ensemble of Neural Networks Utilizing Seismic Attributes for Rock‐Property Inversion With Uncertainty Estimation

Contribution to advancing aquifer geometric mapping using machine learning and deep learning techniques: a case study of the AL Haouz-Mejjate aquifer, Marrakech, Morocco

Contact Info

Product

Resources

About