Joint Prediction Model of Reservoir Parameters Based on Multimodal Transformer Graph Neural Operator Physical Constraint Network

Yanyu Ding; Guoqing Chen

doi:10.71451/ISTAER2604

Authors

Yanyu Ding School of Finance, Chengdu Jincheng College, Chengdu, Sichuan, China Author https://orcid.org/0009-0002-9922-8752
Guoqing Chen School of Finance, Chengdu Jincheng College, Chengdu, Sichuan, China Author https://orcid.org/0000-0002-1483-7720

DOI:

https://doi.org/10.71451/ISTAER2604

Keywords:

Reservoir parameter prediction; Multimodal transformer; Graph neural operator; Physical constraint network; Multi source data fusion

Abstract

Accurate prediction of reservoir parameters is the core of reservoir evaluation and development plan optimization, but traditional methods are difficult to effectively integrate multi-source heterogeneous data, depict complex spatial heterogeneity and ensure physical consistency. Therefore, this paper proposes a multimodal transformer graph neural operator physical constraint network (MT-GNO-PCN) to realize the joint high-precision prediction of reservoir parameters. Firstly, the multimodal transformer is used to integrate seismic attributes, logging curves and geological interpretation data to construct a unified semantic feature representation; Then the map neural operator is used to learn the continuous space mapping function and flexibly model the distribution law of reservoir parameters in irregular geometry and complex geological structure; Finally, the physical constraint loss term based on the relationship between Darcy's law and rock physics is introduced to enhance the physical rationality and generalization ability of the prediction results. Experiments on real and synthetic reservoir data sets show that the average mean square error of this method is about 46% lower than that of the traditional convolutional neural network and 33% lower than that of the model using only multimodal transformer in the prediction of porosity, permeability and water saturation; The average determination coefficient (R²) is 0.89, and the error increase is controlled within 165% under 15% noise interference, which is significantly better than the existing comparison model. The framework provides a new way for reservoir intelligent modeling driven by multi-source data with high accuracy, strong robustness and physical consistency.

References

[1] Khalili, Y., & Ahmadi, M. (2023). Reservoir Modeling & Simulation: Advancements, Challenges, and Future Perspectives. Journal of Chemical & Petroleum Engineering, 57(2).

[2] Aljehani, A. S. (2025). Artificial intelligence for reservoir modeling and property estimation in petroleum engineering. Physics and Chemistry of the Earth, Parts A/B/C, 140, 104015. DOI: https://doi.org/10.1016/j.pce.2025.104015

[3] Uchendu, O., Omomo, K. O., & Esiri, A. E. (2024). Conceptual framework for data-driven reservoir characterization: Integrating machine learning in petrophysical analysis. Comprehensive Research and Reviews in Multidisciplinary Studies, 2(2), 1-13. DOI: https://doi.org/10.57219/crrms.2024.2.2.0041

[4] Ehsan, M., Chen, R., Abdelrahman, K., Manzoor, U., Hussain, M., Ullah, J., & Zaheer, A. M. (2025). Application of Petrophysical Analysis, Rock Physics, Seismic Attributes, Seismic Inversion, Multi-attribute Analysis, and Probabilistic Neural Networks for Estimating Petrophysical Parameters for Source and Reservoir Rock Evaluations in the Lower Indus Basin, Pakistan: M. Ehsan et al. Natural Resources Research, 34(6), 3073-3101. DOI: https://doi.org/10.1007/s11053-025-10550-6

[5] Grana, D., Azevedo, L., De Figueiredo, L., Connolly, P., & Mukerji, T. (2022). Probabilistic inversion of seismic data for reservoir petrophysical characterization: Review and examples. Geophysics, 87(5), M199-M216. DOI: https://doi.org/10.1190/geo2021-0776.1

[6] Cao, X., Liu, Z., Hu, C., Song, X., Quaye, J. A., & Lu, N. (2024). Three-dimensional geological modelling in earth science research: an in-depth review and perspective analysis. Minerals, 14(7), 686. DOI: https://doi.org/10.3390/min14070686

[7] Zhao, T., Wang, S., Ouyang, C., Chen, M., Liu, C., Zhang, J., ... & Wang, L. (2024). Artificial intelligence for geoscience: Progress, challenges, and perspectives. The Innovation, 5(5). DOI: https://doi.org/10.1016/j.xinn.2024.100691

[8] Zhang, L., Xie, Y., Xidao, L., & Zhang, X. (2018, May). Multi-source heterogeneous data fusion. In 2018 International conference on artificial intelligence and big data (ICAIBD) (pp. 47-51). IEEE. DOI: https://doi.org/10.1109/ICAIBD.2018.8396165

[9] Zhao, X., Jia, Y., Li, A., Jiang, R., & Song, Y. (2020). Multi-source knowledge fusion: a survey. World Wide Web, 23(4), 2567-2592. DOI: https://doi.org/10.1007/s11280-020-00811-0

[10] Liu, Y., Zhang, G., Hao, J., & Chen, Y. (2025). Multi-source knowledge fusion framework: assessing suppliers’ research and development of complex products. Management Decision. DOI: https://doi.org/10.1108/MD-11-2024-2546

[11] Wang, B., Wu, L., Li, W., Qiu, Q., Xie, Z., Liu, H., & Zhou, Y. (2021). A semi-automatic approach for generating geological profiles by integrating multi-source data. Ore Geology Reviews, 134, 104190. DOI: https://doi.org/10.1016/j.oregeorev.2021.104190

[12] Zhang, Q., Zhu, L., & Huang, D. S. (2018). High-order convolutional neural network architecture for predicting DNA-protein binding sites. IEEE/ACM transactions on computational biology and bioinformatics, 16(4), 1184-1192. DOI: https://doi.org/10.1109/TCBB.2018.2819660

[13] Liu, B., Tang, R., Chen, Y., Yu, J., Guo, H., & Zhang, Y. (2019, May). Feature generation by convolutional neural network for click-through rate prediction. In The World Wide Web Conference (pp. 1119-1129). DOI: https://doi.org/10.1145/3308558.3313497

[14] Wu, K., Liu, J., Liu, P., & Yang, S. (2019). Time series prediction using sparse autoencoder and high-order fuzzy cognitive maps. IEEE transactions on fuzzy systems, 28(12), 3110-3121. DOI: https://doi.org/10.1109/TFUZZ.2019.2956904

[15] Li, X., Qiu, Y., Zhou, J., & Xie, Z. (2021). Applications and challenges of machine learning methods in Alzheimer's disease multi-source data analysis. Current Genomics, 22(8), 564-582. DOI: https://doi.org/10.2174/1389202923666211216163049

[16] Jiang, Y., Li, C., Sun, L., Guo, D., Zhang, Y., & Wang, W. (2021). A deep learning algorithm for multi-source data fusion to predict water quality of urban sewer networks. Journal of Cleaner Production, 318, 128533. DOI: https://doi.org/10.1016/j.jclepro.2021.128533

[17] Guo, D., Yang, X., Peng, P., Zhu, L., & He, H. (2025). The intelligent fault identification method based on multi-source information fusion and deep learning. Scientific Reports, 15(1), 6643. DOI: https://doi.org/10.1038/s41598-025-90823-5

[18] Liu, Y., Liao, S., Yang, Y., & Zhang, B. (2024). Data-driven and physics-informed neural network for predicting tunnelling-induced ground deformation with sparse data of field measurement. Tunnelling and Underground Space Technology, 152, 105951. DOI: https://doi.org/10.1016/j.tust.2024.105951

[19] Taloma, R. J. L., Cuomo, F., Comminiello, D., & Pisani, P. (2025). Machine learning for smart water distribution systems: exploring applications, challenges and future perspectives. Artificial Intelligence Review, 58(4), 120. DOI: https://doi.org/10.1007/s10462-024-11093-7

[20] Qiang, Z., Yasin, Q., Golsanami, N., & Du, Q. (2020). Prediction of reservoir quality from log-core and seismic inversion analysis with an artificial neural network: A case study from the Sawan Gas Field, Pakistan. Energies, 13(2), 486. DOI: https://doi.org/10.3390/en13020486

[21] Gogoi, T., & Chatterjee, R. (2019). Estimation of petrophysical parameters using seismic inversion and neural network modeling in Upper Assam basin, India. Geoscience Frontiers, 10(3), 1113-1124. DOI: https://doi.org/10.1016/j.gsf.2018.07.002

[22] Bashir, Y., Siddiqui, N. A., Morib, D. L., Babasafari, A. A., Ali, S. H., Imran, Q. S., & Karaman, A. (2024). Cohesive approach for determining porosity and P-impedance in carbonate rocks using seismic attributes and inversion analysis. Journal of Petroleum Exploration and Production Technology, 14(5), 1173-1187. DOI: https://doi.org/10.1007/s13202-024-01767-x

[23] Rostami, A., Kordavani, A., Parchekhari, S., Hemmati-Sarapardeh, A., & Helalizadeh, A. (2022). New insights into permeability determination by coupling Stoneley wave propagation and conventional petrophysical logs in carbonate oil reservoirs. Scientific Reports, 12(1), 11618. DOI: https://doi.org/10.1038/s41598-022-15869-1

[24] Agbadze, O. K., Qiang, C., & Jiaren, Y. (2022). Acoustic impedance and lithology-based reservoir porosity analysis using predictive machine learning algorithms. Journal of Petroleum Science and Engineering, 208, 109656. DOI: https://doi.org/10.1016/j.petrol.2021.109656

[25] Riahi, M. A., & Fakhari, M. G. (2022). Pore pressure prediction using seismic acoustic impedance in an overpressure carbonate reservoir. Journal of Petroleum Exploration and Production Technology, 12(12), 3311-3323. DOI: https://doi.org/10.1007/s13202-022-01524-y

[26] Lang, X., Li, C., Wang, M., & Li, X. (2024). Semi-supervised seismic impedance inversion with convolutional neural network and lightweight transformer. IEEE Transactions on Geoscience and Remote Sensing, 62, 1-11. DOI: https://doi.org/10.1109/TGRS.2024.3401225

[27] Farahani, S., & Bahroudi, A. (2025). 3D mineral prospectivity modeling using a multi-scale CNN–transformer: A case study from the Siahcheshmeh gold deposit, NW Iran. Ore Geology Reviews, 107066. DOI: https://doi.org/10.1016/j.oregeorev.2025.107066

[28] Wang, J., Yu, L., & Tian, S. (2025). Cross-attention interaction learning network for multi-model image fusion via transformer. Engineering Applications of Artificial Intelligence, 139, 109583. DOI: https://doi.org/10.1016/j.engappai.2024.109583

[29] Zheng, F., Li, W., Wang, X., Wang, L., Zhang, X., & Zhang, H. (2022). A cross-attention mechanism based on regional-level semantic features of images for cross-modal text-image retrieval in remote sensing. Applied Sciences, 12(23), 12221. DOI: https://doi.org/10.3390/app122312221

[30] Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2020). A comprehensive survey on graph neural networks. IEEE transactions on neural networks and learning systems, 32(1), 4-24. DOI: https://doi.org/10.1109/TNNLS.2020.2978386

[31] Du, Y., Chen, G., Pang, C., & Zhao, T. (2026). Prediction of fracture and vug parameters in carbonate reservoirs using a combined T-GNO-PINN approach. Journal of Seismic Exploration, 35(1), Article 025330057. DOI: https://doi.org/10.36922/JSE025330057

[32] Chen, G., Zhao, T., Pang, C., Seenoi, P., Papukdee, N., & Busababodhin, P. (2025). An attention-guided graph neural network and U-Net++-based reservoir porosity prediction system. Journal of Seismic Exploration, 34(4), 70-87. DOI: https://doi.org/10.36922/JSE025300044