A data-driven model to estimate the pore volume to breakthrough for carbonate acidizing

Abstract

This research investigates the impact of different rock, acid, and reaction dynamic properties on the pore volume to breakthough (PVBT) at different acid injection rates using in-house developed two-scale continuum simulation model. We analyzed the parameters relation and developed a reliable machine learning model to accurately predict the PVBT at similar range of investigated parameters. In the simulation, it was found that different acid concentrations result in the same optimum injection velocity but at large contrast in PVBT between low and high acid concentration. However, other parameters such as diffusion coefficient and reaction rate exhibited an inverse PVBT behavior across optimum injection velocity due to change in acid transport and reaction behavior. After that, different reliable machine learning algorithms were employed to predict the optimum PVBT for carbonates matrix acidizing. The utilized machine learning models undergone multiple optimizations and comparison to obtain the most accurate prediction performance. The artificial neural network model with 2 hidden layers outperforms the other optimizations with 11.27% estimation error, 0.96 R2 and 0.98 correlation coefficient for the testing data set. Finally, an empirical correlation was developed to accurately estimate PVBT at a low cost and very short time compared to lab experiments and numerical simulation models. The novelty of this research stems from examining PVBT curves behavior by varying five matrix acidizing parameters independently, analyzing the correlation between these parameters and developing machine learning model for handy and reliable optimum PVBT estimation.

Optimizing acidizing design and effectiveness assessment with machine learning for predicting post-acidizing permeability

“…XGBoost enhances performance and reduces overfitting, whereas random forest combines decision trees for robust predictions. Overall, these models were chosen due to their capabilities in handling the complexities of acidizing and their track record of accurate predictions 17 , 24 , 25 , 28 , 30 , 34 , 35 , 44 .…”

“…Gumrah et al describe a computer model that uses a genetic algorithm to optimize Damkohler and acid capacity numbers for predicting the permeability alteration of an acidization process 31 – 33 . Alkathim et al investigated the impact of rock, acid, and reaction properties on pore volume to breakthrough during calcite matrix acidizing, finding optimal injection rates 34 , while Kurniawan proposed a machine learning and regression analysis model to enhance success rates and net oil gain in hydraulic fractured sandstone formations, improving candidate selection 35 . Additionally, Abdollah Hatamizadeh and Behnam Sedaee optimized acidizing processes in carbonate reservoirs using neural networks, meta-learning algorithms, and genetic algorithms, achieving high simulation accuracy and minimizing acid consumption while enhancing permeability improvement 17 .…”

Machine Learning in Reservoir Engineering: A Review

“…Prediction of optimal pumping rate SVM [15] 2018 Prediction of pore volume breakthrough (PVBT) GA [16] 2019; ANN [17] 2023…”

“…This method was validated with 170 experimental data collected from Indiana, Desert Rose, and Law Limestone and was able to predict them with 90% accuracy. Alkathim et al [17] found that different acid concentrations, diffusion coefficients, and reaction rates lead to significant differences in PVBT. They used an ANN to predict the optimal PVBT in carbonate acidization.…”

“…In the fields of hydraulic fracturing and acidizing, chemical flooding and gas flooding, and water injection in reservoir engineering, the application of machine learning can be summarized as numerical prediction. The application of machine learning in these scenarios mainly involves the following steps [17,36,55].…”