Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm

Unconventional reservoirs are characterized by intricate fluid-phase behaviors, and physics-based shale oil well productivity prediction models often exhibit substantial deviations due to oversimplified theoretical frameworks and challenges in parameter acquisition. Under these circumstances, data-d...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuan Pan, Xuewei Liu, Fuchun Tian, Liyong Yang, Xiaoting Gou, Yunpeng Jia, Quan Wang, Yingxi Zhang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/18/10/2523
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849710893469794304
author Yuan Pan
Xuewei Liu
Fuchun Tian
Liyong Yang
Xiaoting Gou
Yunpeng Jia
Quan Wang
Yingxi Zhang
author_facet Yuan Pan
Xuewei Liu
Fuchun Tian
Liyong Yang
Xiaoting Gou
Yunpeng Jia
Quan Wang
Yingxi Zhang
author_sort Yuan Pan
collection DOAJ
description Unconventional reservoirs are characterized by intricate fluid-phase behaviors, and physics-based shale oil well productivity prediction models often exhibit substantial deviations due to oversimplified theoretical frameworks and challenges in parameter acquisition. Under these circumstances, data-driven approaches leveraging actual production datasets have emerged as viable alternatives for productivity forecasting. Nevertheless, conventional data-driven architectures suffer from structural simplicity, limited capacity for processing low-dimensional feature spaces, and exclusive applicability to intra-sequence learning paradigms (e.g., production-to-production sequence mapping). This fundamentally conflicts with the underlying principles of mechanistic modeling, which emphasize pressure-to-production sequence transformations. To address these limitations, we propose a hybrid deep learning architecture integrating convolutional neural networks with bidirectional gated recurrent units (CNN-BiGRU). The model incorporates dedicated input pathways: fully connected layers for feature embedding and convolutional operations for high-dimensional feature extraction. By implementing a sequence-to-sequence (seq2seq) architecture with encoder–decoder mechanisms, our framework enables cross-domain sequence learning, effectively bridging pressure dynamics with production profiles. The CNN-BiGRU model was implemented on the TensorFlow framework, with rigorous validation of model robustness and systematic evaluation of feature importance. Hyperparameter optimization via grid searching yielded optimal configurations, while field applications demonstrated operational feasibility. Comparative analysis revealed a mean relative error (MRE) of 16.11% between predicted and observed production values, substantiating the model’s predictive competence. This methodology establishes a novel paradigm for machine learning-driven productivity prediction in unconventional reservoir engineering.
format Article
id doaj-art-341556f35ffb4520a9426da5145366e9
institution DOAJ
issn 1996-1073
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-341556f35ffb4520a9426da5145366e92025-08-20T03:14:46ZengMDPI AGEnergies1996-10732025-05-011810252310.3390/en18102523Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU AlgorithmYuan Pan0Xuewei Liu1Fuchun Tian2Liyong Yang3Xiaoting Gou4Yunpeng Jia5Quan Wang6Yingxi Zhang7Petroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaPetroleum Engineering Research Institute, PetroChina Dagang Oilfield Company, Tianjin 300280, ChinaUnconventional reservoirs are characterized by intricate fluid-phase behaviors, and physics-based shale oil well productivity prediction models often exhibit substantial deviations due to oversimplified theoretical frameworks and challenges in parameter acquisition. Under these circumstances, data-driven approaches leveraging actual production datasets have emerged as viable alternatives for productivity forecasting. Nevertheless, conventional data-driven architectures suffer from structural simplicity, limited capacity for processing low-dimensional feature spaces, and exclusive applicability to intra-sequence learning paradigms (e.g., production-to-production sequence mapping). This fundamentally conflicts with the underlying principles of mechanistic modeling, which emphasize pressure-to-production sequence transformations. To address these limitations, we propose a hybrid deep learning architecture integrating convolutional neural networks with bidirectional gated recurrent units (CNN-BiGRU). The model incorporates dedicated input pathways: fully connected layers for feature embedding and convolutional operations for high-dimensional feature extraction. By implementing a sequence-to-sequence (seq2seq) architecture with encoder–decoder mechanisms, our framework enables cross-domain sequence learning, effectively bridging pressure dynamics with production profiles. The CNN-BiGRU model was implemented on the TensorFlow framework, with rigorous validation of model robustness and systematic evaluation of feature importance. Hyperparameter optimization via grid searching yielded optimal configurations, while field applications demonstrated operational feasibility. Comparative analysis revealed a mean relative error (MRE) of 16.11% between predicted and observed production values, substantiating the model’s predictive competence. This methodology establishes a novel paradigm for machine learning-driven productivity prediction in unconventional reservoir engineering.https://www.mdpi.com/1996-1073/18/10/2523production predictionneural network deep learningshale oil reservoirsfracturing horizontal wells
spellingShingle Yuan Pan
Xuewei Liu
Fuchun Tian
Liyong Yang
Xiaoting Gou
Yunpeng Jia
Quan Wang
Yingxi Zhang
Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
Energies
production prediction
neural network deep learning
shale oil reservoirs
fracturing horizontal wells
title Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
title_full Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
title_fullStr Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
title_full_unstemmed Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
title_short Research on Shale Oil Well Productivity Prediction Model Based on CNN-BiGRU Algorithm
title_sort research on shale oil well productivity prediction model based on cnn bigru algorithm
topic production prediction
neural network deep learning
shale oil reservoirs
fracturing horizontal wells
url https://www.mdpi.com/1996-1073/18/10/2523
work_keys_str_mv AT yuanpan researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT xueweiliu researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT fuchuntian researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT liyongyang researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT xiaotinggou researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT yunpengjia researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT quanwang researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm
AT yingxizhang researchonshaleoilwellproductivitypredictionmodelbasedoncnnbigrualgorithm