Recently, the team of Zhang Kai has made new progress in the field of characterization of injection-production inter-well connectivity characterization, and the related research has been published in Mathematics. The paper is titled " Reservoir Characterization and Productivity Forecast Based on Knowledge Interaction Neural Network ".
Innovation: In this paper, we have embedded the physical knowledge into neural networks to solve the reservoir characterization and production forecast problems. Integrating the material balance equation with the machine learning techniques, the physical knowledge interaction neural networks have been proposed, enhancing both the merits of interpretability and robustness. Furthermore, the proposed gate functions have avoided the negative connectivity values without physical sense, and the computation efficiency has been fully improved by unconstraint optimization algorithm. In the end, the effectiveness of our models has been proved through several simulation experiments. Also, the performance of the proposed models on noisy data has been demonstrated. KINN illustrates a novel configuration to realize the cooperation and interaction between neural networks and physical knowledge. In the future, we would like to extend KINN to other areas, like production optimization.
Abstract: The reservoir characterization aims to provide the analysis and quantification of the injection-production relationship, which is the fundamental work for production management. The connectivity between injectors and producers is dominated by geological properties, especially permeability. However, the permeability parameters are very heterogenous in oil reservoirs, and expensive to collect by well logging. The commercial simulators enable to get accurate simulation but require sufficient geological properties and consume excessive computation re-sources. In contrast, the data-driven models (physical models and machine learning models) are developed on the observed dynamic data, such as the rate and pressure data of the injectors and producers, constructing the connectivity relationship and forecasting the productivity by a series of nonlinear mappings or the control of specific physical principles. While, due to the “black box” feature of machine learning approaches, and the constraints and assumptions of physical models, the data-driven methods often face the challenges of poor interpretability and generalizability and the limited application scopes. To solve these issues, integrating the physical principle of the waterflooding process (material balance equation) with an artificial neural network (ANN), a knowledge interaction neural network (KINN) is proposed. KINN consists of three transparent modules with explicit physical significance, and different modules are joined together via the material balance equation and work cooperatively to approximate the waterflooding process. In addition, a gate function is proposed to distinguish the dominant flowing channels from weak connecting ones by their sparsity, and thus the inter-well connectivity can be indicated directly by the model parameters. Combining the strong nonlinear mapping ability with the guidance of physical knowledge, the interpretability of KINN is fully enhanced, and the prediction accuracy on the well productivity is improved. The effectiveness of KINN is proved by comparing its performance with the canonical ANN, on the inter-well connectivity analysis and productivity forecast tasks of three synthetic reservoir experiments. Meanwhile, the robustness of KINN is revealed by the sensitivity analysis on measurement noises and wells shut-in cases.
Mathematics covers the fields of mathematical sciences, including timely and thorough survey articles on current trends, new theoretical techniques, novel ideas and new mathematical tools in different branches of mathematics. Topics include: algebraic geometry; algebraic topology; analysis of PDEs; complex variables; differential geometry; dynamical systems; functional analysis; geometric topology; mathematical physics; metric geometry; optimization and control; spectral theory; statistics theory; variational problems; mathematical finance; quantum theory; mathematical and computational biology, etc. The latest impact factor of the journal is 2.592, and the average impact factor IF in the past 5 years is 2.542. This journal currently appears in the JCR Q1 or the Chinese Academy of Sciences ranking 3.
Paper link:
https://doi.org/10.3390/math10091614
Citation:
Jiang Y, Zhang H, Zhang K, et al. Reservoir Characterization and Productivity Forecast Based on Knowledge Interaction Neural Network [J]. Mathematics, 2022, 10(9): 1614.
