Harnessing Advanced Machine-Learning Algorithms for Optimized Data Conditioning and Petrophysical Analysis of Heterogeneous, Thin Reservoirs

Abstract

Petrophysical analysis is an industry-standard practice for reservoir evaluation as it provides critical inputs for characterizing subsurface formations and estimating resource potential. Khadro/Ranikot Formation sands are proliferous producers in the Central Indus Basin, Pakistan. The demarcate potential in intercalated sand shale layers that are thin and heterogeneous makes it a challenging reservoir. Conventional petrophysical interpretation is laborious and does not produce upto-mark results due to reservoir complexity, data limitations, and associated uncertainties. Hence, an emerging and delicate machinelearning (ML) approach has been comprehensively applied to analyze the potential and robustly interpret well log data while addressing the associated challenges. This case study entails a thorough evaluation of well log quality, assessing several algorithms such as least-squares support vector machines (one-class SVM), Random Forest Regressor (RFR), Extra Tree Regressor (ETR), Gradient Boosting Regressor (GBR), Decision Tree Classifier (DTC), etc. to compare their efficacy and reliability. One-class SVM helps to reduce outliers with great certainty, while the missing logs sonic (DT) and density (RHOB) are precisely predicted via GBR and ETR with 0.66 and 0.88 R 2 , respectively. Hence, providing reliable and optimized quality logs suitable for ML-based petrophysics. ML worked on these augmented logs by dividing the data into 60% training and 40% testing. The ETR outperformed the rest of the models with a correlation of 0.99 and 0.91 among conventional and ML results. Likewise, RFR performed exceptionally well for water saturation modeling, expressing the highest 0.93 correlation. Finally, DTC modeled reservoir facies with the best 91% accuracy and 0.935 F1 measures at the blind well. Excellent calibration of >85% is met with the estimates obtained by the predictive model compared to conventional methods. This comprehensive approach offers cost-effective and robust workarounds for modern formation evaluation with minimal uncertainty and resource-efficient multiwell interpretation within complex reservoirs and sets the stage for further research in the machine-learning ecosystem.

ResNet and CWT Fusion: A New Paradigm for Optimized Heterogeneous Thin Reservoir Evaluation

“…The seismic data’s vertical resolution is less than potential sand bed thicknesses, which is determined by the λ/4th i.e., one-fourth of the trace wavelength. 49 The observed seismic data peak frequency is 20 Hz and the reservoir sand interval velocity is 2650 m/s, hence depicting a maximum resolution up to 47 m, proving the tuning of thin-bedded potential sands at various locations throughout the field. 50 Hence, band-limited seismic along with its inverted properties with limited vertical resolution would be incapable of producing credible results for comprehensive reservoir characterization.…”

“…Starting with the analysis of well logs and seismic data, a reliable seismic correlation is established and three horizons from top to bottom, i.e., LGF, Badin shale, and Sand below Badin shale, are subsequently mapped to determine their spatial extent. The seismic data’s vertical resolution is less than potential sand bed thicknesses, which is determined by the λ/4th i.e., one-fourth of the trace wavelength . The observed seismic data peak frequency is 20 Hz and the reservoir sand interval velocity is 2650 m/s, hence depicting a maximum resolution up to 47 m, proving the tuning of thin-bedded potential sands at various locations throughout the field .…”

“…The modeling of corresponding logs due to washouts along with measuring of logs in splice zones is a fundamental step in the characterization of rock strata employed in petrophysical properties estimation. 49 The modeling of missing logs was carried out after the removal of outliers and qualitative review of input data. In the present research, the SML technique was adopted for improving the measured log models.…”

Pore Pressure Prediction for High-Pressure Tight Sandstone in the Huizhou Sag, Pearl River Mouth Basin, China: A Machine Learning-Based Approach

“…∆t m is the transit time in the shale matrix, ∆t ml is the transit time at the mudline (Z = 0), (Z is the true vertical depth below the mudline, (c is the compaction parameter, σe is the effective stress [46], p f is the pore pressure, p pg is the pore pressure gradient, σv g σ is the overburden pressure, p hg is the hydrostatic pressure gradient, and P pg is the formation pore pressure gradient [16,47].…”

Comprehensive Review of Hydrogen Leakage in Relation to Fuel Cell Vehicles and Hydrogen Refueling Stations: Status, Challenges, and Future Prospects

“…Following this, an appropriate algorithm is selected based on the characteristics of the leakage problem to establish and train a model. This model is divided into two parts: one for performance evaluation using a validation data set and the other for regular updates and optimization using newly collected data and feedback . Finally, the trained model is deployed on actual equipment for real-time analysis of sensor data, accurately pinpointing the location and severity of the leak.…”