Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network

The accurate forecasting of crude oil production in CCUS-EOR (carbon capture, utilization, and storage–enhanced oil recovery) operations is essential for the economic evaluation and production optimization of oilfield blocks. Although numerous deep learning models have been widely applied for this p...

Full description

Saved in:
Bibliographic Details
Main Authors: Wei Xia, Qiu Li, Quan Shi, Rui Xu, Jiangtao Wu, Song Deng
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/18/11/2795
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722554541932544
author Wei Xia
Qiu Li
Quan Shi
Rui Xu
Jiangtao Wu
Song Deng
author_facet Wei Xia
Qiu Li
Quan Shi
Rui Xu
Jiangtao Wu
Song Deng
author_sort Wei Xia
collection DOAJ
description The accurate forecasting of crude oil production in CCUS-EOR (carbon capture, utilization, and storage–enhanced oil recovery) operations is essential for the economic evaluation and production optimization of oilfield blocks. Although numerous deep learning models have been widely applied for this purpose, existing methods still face challenges in extracting complex features from multidimensional time series datasets, limiting the accuracy of oil production forecasts. In this study, we propose a novel KAN-LSTM model that integrates a KAN (knowledge-aware network) layer with a long short-term memory (LSTM) neural network to enhance the accuracy of oil production forecasting in CCUS-EOR applications. The KAN layer effectively extracts relevant features from multivariate data, while the LSTM layer models temporal information based on the extracted features to generate accurate predictions. To evaluate the performance of the proposed model, we conducted two case studies using both mechanistic model data and real project production data. The prediction performance of our method was compared with that of typical deep learning approaches. Experimental results demonstrate that the KAN-LSTM model outperforms other forecasting methods. By providing reliable estimates of future oil production, the KAN-LSTM model enables engineers to make informed decisions in reservoir development planning.
format Article
id doaj-art-c3a09ae83446444a81ec9cfbaa096e46
institution DOAJ
issn 1996-1073
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-c3a09ae83446444a81ec9cfbaa096e462025-08-20T03:11:18ZengMDPI AGEnergies1996-10732025-05-011811279510.3390/en18112795Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural NetworkWei Xia0Qiu Li1Quan Shi2Rui Xu3Jiangtao Wu4Song Deng5Key Laboratory of Thermo-Fluid Science and Engineering, Ministry of Education, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaSchool of Petroleum and Natural Gas Engineering, Changzhou University, Changzhou 213164, ChinaSchool of Petroleum and Natural Gas Engineering, Changzhou University, Changzhou 213164, ChinaZhejiang Oilfield Company, PetroChina Company Limited, Hangzhou 310000, ChinaKey Laboratory of Thermo-Fluid Science and Engineering, Ministry of Education, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an 710049, ChinaSchool of Petroleum and Natural Gas Engineering, Changzhou University, Changzhou 213164, ChinaThe accurate forecasting of crude oil production in CCUS-EOR (carbon capture, utilization, and storage–enhanced oil recovery) operations is essential for the economic evaluation and production optimization of oilfield blocks. Although numerous deep learning models have been widely applied for this purpose, existing methods still face challenges in extracting complex features from multidimensional time series datasets, limiting the accuracy of oil production forecasts. In this study, we propose a novel KAN-LSTM model that integrates a KAN (knowledge-aware network) layer with a long short-term memory (LSTM) neural network to enhance the accuracy of oil production forecasting in CCUS-EOR applications. The KAN layer effectively extracts relevant features from multivariate data, while the LSTM layer models temporal information based on the extracted features to generate accurate predictions. To evaluate the performance of the proposed model, we conducted two case studies using both mechanistic model data and real project production data. The prediction performance of our method was compared with that of typical deep learning approaches. Experimental results demonstrate that the KAN-LSTM model outperforms other forecasting methods. By providing reliable estimates of future oil production, the KAN-LSTM model enables engineers to make informed decisions in reservoir development planning.https://www.mdpi.com/1996-1073/18/11/2795CCUS-EORKAN-LSTMdeep learning
spellingShingle Wei Xia
Qiu Li
Quan Shi
Rui Xu
Jiangtao Wu
Song Deng
Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
Energies
CCUS-EOR
KAN-LSTM
deep learning
title Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
title_full Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
title_fullStr Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
title_full_unstemmed Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
title_short Enhanced Oil Production Forecasting in CCUS-EOR Systems via KAN-LSTM Neural Network
title_sort enhanced oil production forecasting in ccus eor systems via kan lstm neural network
topic CCUS-EOR
KAN-LSTM
deep learning
url https://www.mdpi.com/1996-1073/18/11/2795
work_keys_str_mv AT weixia enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork
AT qiuli enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork
AT quanshi enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork
AT ruixu enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork
AT jiangtaowu enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork
AT songdeng enhancedoilproductionforecastinginccuseorsystemsviakanlstmneuralnetwork