Fuyong Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Fuyong Wang has a total of 138 co-authors. Here are the top co-authors by number of shared publications:
Yiqiang Li
China University of Petroleum, Beijing
12 shared publications- Hu Guo
China University of Petroleum-Beijing
11 shared publications Jiuyu Zhao
10 shared publications- Hui Cheng
Key Laboratory of Groundwater Resources and Environment, Ministry of Education, Jilin University 1 , Changchun 130012
8 shared publications Liang Jiao
Unconventional Petroleum Research Institute, China University of Petroleum, Beijing 102249, China
7 shared publicationsZhichao Liu
Unconventional Petroleum Research Institute, China University of Petroleum, Beijing 102249, China
7 shared publications
Latest Publications
- 1 August 2026
- Published by AIP Publishing in Physics of Fluids
- Vol. 38 (8)
Dynamic inter-well connectivity inversion for waterflooded reservoirs using long short-term memory and graph neural networks with genetic algorithm optimization
Accurate quantification of inter-well connectivity is critical for optimizing recovery in waterflooded reservoirs at the high water-cut stage. Conventional approaches often fail to capture the complex spatiotemporal dependencies in large-scale well patterns, particularly under dynamic flow with pronounced injection–production time delays. To address these challenges, this study proposes a novel data-driven framework, termed LSTM-GAGNN, which integrates long short-term memory (LSTM) networks, graph neural networks (GNN), and genetic algorithm (GA). In the proposed approach, LSTM networks are first employed to extract long-term temporal dependencies from historical production and injection data, effectively accounting for lagged responses. The well system is then represented as a graph structure, enabling the GNN model to capture spatial interactions and dynamically learn inter-well relationships. Finally, a GA is incorporated to perform global optimization of model parameters, thereby enhancing solution robustness and ensuring the physical interpretability of the inferred connectivity matrix. The proposed framework is validated using numerical reservoir simulation cases and further tested on field data from three well patterns in an eastern China oilfield. The results demonstrate that the LSTM-GAGNN model can accurately and robustly characterize dynamic inter-well connectivity, outperforming conventional methods in both stability and predictive capability. This work provides a reliable and scalable data-driven approach for connectivity inversion in complex reservoirs, offering valuable support for intelligent reservoir management, proactive injection–production optimization, and improved field development decision-making.
- 1 August 2026
- Published by AIP Publishing in Physics of Fluids
- Vol. 38 (8)
Dynamic inter-well connectivity inversion for waterflooded reservoirs using long short-term memory and graph neural networks with genetic algorithm optimization
Accurate quantification of inter-well connectivity is critical for optimizing recovery in waterflooded reservoirs at the high water-cut stage. Conventional approaches often fail to capture the complex spatiotemporal dependencies in large-scale well patterns, particularly under dynamic flow with pronounced injection–production time delays. To address these challenges, this study proposes a novel data-driven framework, termed LSTM-GAGNN, which integrates long short-term memory (LSTM) networks, graph neural networks (GNN), and genetic algorithm (GA). In the proposed approach, LSTM networks are first employed to extract long-term temporal dependencies from historical production and injection data, effectively accounting for lagged responses. The well system is then represented as a graph structure, enabling the GNN model to capture spatial interactions and dynamically learn inter-well relationships. Finally, a GA is incorporated to perform global optimization of model parameters, thereby enhancing solution robustness and ensuring the physical interpretability of the inferred connectivity matrix. The proposed framework is validated using numerical reservoir simulation cases and further tested on field data from three well patterns in an eastern China oilfield. The results demonstrate that the LSTM-GAGNN model can accurately and robustly characterize dynamic inter-well connectivity, outperforming conventional methods in both stability and predictive capability. This work provides a reliable and scalable data-driven approach for connectivity inversion in complex reservoirs, offering valuable support for intelligent reservoir management, proactive injection–production optimization, and improved field development decision-making.
- Open Access
- 1 December 2025
- Published by KEAI Publishing in Petroleum Research
- Vol. 10 (4) , 699-709
Production forecasting in carbonate oil reservoirs through integration of deep learning and geological parameters
- 1 December 2025
- Published by KEAI Publishing in Petroleum Research
- Vol. 10 (4) , 699-709
Open AccessProduction forecasting in carbonate oil reservoirs through integration of deep learning and geological parameters
- Open Access
- 2 June 2025
- Published by American Chemical Society (ACS) in ACS Omega
- Vol. 10 (22) , 23011-23024
Reservoir Flow Field Characterization during Water Flooding: A Data Mining and Fuzzy Logic Approach
Accurate characterization of reservoir flow fields is critical for enhancing oil recovery in water flooding oil fields. Current methodologies lack standardized criteria for selecting influential factors in flow field characterization. This study introduces a novel quantitative framework integrating data mining and fuzzy logic to address this gap. Principal component analysis (PCA) and Spearman’s rank correlation coefficient are applied to identify key indices, forming a robust flow field characterization system. A dual subjective–objective weighting strategy, combining the analytic hierarchy process (AHP) with entropy weighting, determines comprehensive index weights. Statistical analysis of data distributionincluding quartile deviation, variance, and standard deviation, guides the selection of optimal membership functions for each index. Fuzzy logic is then employed to derive a quantitative characterization formula, while K-means clustering categorizes flow field types. Results indicate that the PCA and Spearman correlation coefficient effectively isolate critical factors: oil surface flux, pressure, oil saturation, and permeability. The weighting strategy assigns greater importance to dynamic factors (0.833) than to static factors (0.167). Membership functions align with observed data distributions, and K-means clustering achieves a silhouette coefficient of 0.724, delineating three distinct flow field types: dominant (10.99%), weak flow field type I (52.71%), and weak flow field type II (36.29%). This data-driven framework provides actionable insights for optimizing production strategies and improving oil recovery in mature, high water-cut oil reservoirs.
- 2 June 2025
- Published by American Chemical Society (ACS) in ACS Omega
- Vol. 10 (22) , 23011-23024
Open AccessReservoir Flow Field Characterization during Water Flooding: A Data Mining and Fuzzy Logic Approach
Accurate characterization of reservoir flow fields is critical for enhancing oil recovery in water flooding oil fields. Current methodologies lack standardized criteria for selecting influential factors in flow field characterization. This study introduces a novel quantitative framework integrating data mining and fuzzy logic to address this gap. Principal component analysis (PCA) and Spearman’s rank correlation coefficient are applied to identify key indices, forming a robust flow field characterization system. A dual subjective–objective weighting strategy, combining the analytic hierarchy process (AHP) with entropy weighting, determines comprehensive index weights. Statistical analysis of data distributionincluding quartile deviation, variance, and standard deviation, guides the selection of optimal membership functions for each index. Fuzzy logic is then employed to derive a quantitative characterization formula, while K-means clustering categorizes flow field types. Results indicate that the PCA and Spearman correlation coefficient effectively isolate critical factors: oil surface flux, pressure, oil saturation, and permeability. The weighting strategy assigns greater importance to dynamic factors (0.833) than to static factors (0.167). Membership functions align with observed data distributions, and K-means clustering achieves a silhouette coefficient of 0.724, delineating three distinct flow field types: dominant (10.99%), weak flow field type I (52.71%), and weak flow field type II (36.29%). This data-driven framework provides actionable insights for optimizing production strategies and improving oil recovery in mature, high water-cut oil reservoirs.
- Top Cited PapersOpen Access
- 1 June 2025
- Published by KEAI Publishing in Energy Geoscience
- Vol. 6 (2)
Machine learning-based prediction of physical parameters in heterogeneous carbonate reservoirs using well log data
- 1 June 2025
- Published by KEAI Publishing in Energy Geoscience
- Vol. 6 (2)
Top Cited PapersOpen AccessMachine learning-based prediction of physical parameters in heterogeneous carbonate reservoirs using well log data
- 20 January 2025
- Published by Elsevier in Geoenergy Science and Engineering
- Vol. 247
A novel approach for production allocation in multi-layer oil reservoirs based on machine learning combining game theory
- 20 January 2025
- Published by Elsevier in Geoenergy Science and Engineering
- Vol. 247
A novel approach for production allocation in multi-layer oil reservoirs based on machine learning combining game theory
- 1 December 2024
- Published by AIP Publishing in Physics of Fluids
- Vol. 36 (12)
A transient analysis method for a multi-horizontal-well pad with well interference and gas–water flow in shale
Complex fracture networks generated by fracturing treatments and the gas–water flow have intensified the well interference in shale gas reservoirs. In this paper, a high-precision numerical model is developed to capture the effects of well interference and gas–water flow on pressure transient behaviors of multi-horizontal-well pads. The unstructured Voronoi grid approach is employed to discretize the physical model. The model's accuracy is verified through published analytical solutions. The findings reveal that a multi-horizontal-well pad in shale gas reservoirs can exhibit up to ten distinct flow regimes. An increase in residual water saturation reduces the water flow ability and leads to an earlier commencement of the gas–water flow regime. Once the formation water becomes flowable, higher water saturation delays the initiation of all flow regimes. The well interference regime exhibits characteristics of the linear flow due to the increased well spacing. As fracture propagation leads to the well communication, the effect of fracture half-length on pressure transient behaviors becomes negligible. The case study from the Fuling shale gas reservoir demonstrates that the proposed model achieves a satisfactory fit to the actual pressure data and the pressure transient behaviors manifest a typical boundary-dominated flow regime.
- 1 December 2024
- Published by AIP Publishing in Physics of Fluids
- Vol. 36 (12)
A transient analysis method for a multi-horizontal-well pad with well interference and gas–water flow in shale
Complex fracture networks generated by fracturing treatments and the gas–water flow have intensified the well interference in shale gas reservoirs. In this paper, a high-precision numerical model is developed to capture the effects of well interference and gas–water flow on pressure transient behaviors of multi-horizontal-well pads. The unstructured Voronoi grid approach is employed to discretize the physical model. The model's accuracy is verified through published analytical solutions. The findings reveal that a multi-horizontal-well pad in shale gas reservoirs can exhibit up to ten distinct flow regimes. An increase in residual water saturation reduces the water flow ability and leads to an earlier commencement of the gas–water flow regime. Once the formation water becomes flowable, higher water saturation delays the initiation of all flow regimes. The well interference regime exhibits characteristics of the linear flow due to the increased well spacing. As fracture propagation leads to the well communication, the effect of fracture half-length on pressure transient behaviors becomes negligible. The case study from the Fuling shale gas reservoir demonstrates that the proposed model achieves a satisfactory fit to the actual pressure data and the pressure transient behaviors manifest a typical boundary-dominated flow regime.
- Open Access
- 1 December 2024
- Published by KEAI Publishing in Journal of Pipeline Science and Engineering
- Vol. 4 (4)
Advances in intelligent identification of fiber-optic vibration signals in oil and gas pipelines
- 1 December 2024
- Published by KEAI Publishing in Journal of Pipeline Science and Engineering
- Vol. 4 (4)
Open AccessAdvances in intelligent identification of fiber-optic vibration signals in oil and gas pipelines
- 13 November 2024
- Published by American Chemical Society (ACS) in Energy & Fuels
- Vol. 38 (22) , 22000-22011
Permeability Prediction and Rock Typing for Unconventional Reservoirs Using High-Pressure Mercury Intrusion and Fractal Analysis
- 13 November 2024
- Published by American Chemical Society (ACS) in Energy & Fuels
- Vol. 38 (22) , 22000-22011
Permeability Prediction and Rock Typing for Unconventional Reservoirs Using High-Pressure Mercury Intrusion and Fractal Analysis