Article Details

Abstract

Oilfield production forecasting is a fundamental task for formulating production and development scheme. With the increasing volume of data, traditional methods face significant challenges. Establishing forecasting models through Machine Learning has become a more accurate and efficient method. This paper combines two fundamental models, Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), to construct a CNN-LSTM model. It extracts some prominent features from the oilfield data by CNN, and uses the features as input to construct the LSTM time series dataset. It reveals the hidden relationship between features and spatial interaction in data, and then enhances the accuracy of the forecasting results. In addition, the experiment results demonstrate that compared with standalone LSTM model and CNN model, the CNN-LSTM model has significant advantages in terms of forecasting performance. It can forecast the oilfield production data more accurately and provide reliable data support for the subsequent oilfield development.

Keywords

CNN-LSTM; Machine Learning; production forecasting

References

1. [1] ZHA W S，LIU Y P，WAN Y J，et al. Forecasting Monthly Gas Field Production Based on the CNN-LSTM Model [J/OL].Energy，2022，260：124889[2024-12-26].https://doi.org/10.1016/j.energy.2022.124889.
2. [2] PAN S W，WANG J Z，ZHOU W. Prediction on Production of Oil Well with Attention-CNN-LSTM [C]//2021 International Conference on Electrical Engineering and Computer Technology.Qingdao：IOP，2021，2030（1）：1-8.
3. [3] ABDULLAYEVA F，IMAMVERDIYEV Y. Development of Oil Production Forecasting Method Based on Deep Learning [J].Statistics，Optimization & Information Computing，2019，7（4）：826-839.
4. [4] Li Liujie, Tang Houjun, Guo Fenzhuan, etc An improved empirical model for predicting horizontal well production in Chang 7 shale oil fields [J]. Unconventional Oil and Gas, 2024, 11 (4): 106-114
5. [5] Zhao Qingjie, Tang Hongbao, Zhang Qian, etc Research on the Prediction Method of Shale Oil Fracturing Well Production [J]. Oil and Gas Well Testing, 2024, 33 (3): 46-52
6. [6] Liu Jiahao, Liu Hao Oil single well production prediction based on LSTM neural network model [J]. Petrochemical Applications, 2023, 42 (3): 38-41+52
7. [7] Wang Xiaolin, Tang Wei, Tang Hongjun, etc Monthly Oil Production Prediction Method Based on Machine Learning [C]//2023 International Conference on Oil and Gas Field Exploration and Development Wuhan: [publisher unknown], 2023
8. [8] Lan Weihua, Li Xinfeng, Deng Zuyue, etc Application and Research of Artificial Intelligence Technology in Offshore Oil and Gas Production [C]//2023 China Oil and Gas Artificial Intelligence Technology Conference Beijing: [Publisher unknown], 2023
9. [9] Teng Qigang Low cost Internet of Things Construction and Application in Jilin Oilfield [J]. Petroleum Science and Technology Forum, 2023, 42 (3): 65-71
10. [10] Wu Xingen, Ge Jiali Application of Artificial Neural Networks for Predicting Oilfield Production [J]. Petroleum Exploration and Development, 1994, 21 (3): 75-78
11. [11] Gao Yajun, Tang Lihui, Wang Zhenpeng, etc Oil field production prediction method based on recurrent neural network and data differential processing [J]. China Offshore Oil and Gas, 2023, 35 (3): 126-136
12. [12] Ren Yanlong, Gu Jianwei Research on BP Neural Network Oilfield Production Prediction Model Based on Genetic Algorithm [C]//The 6th International Academic Conference on Digital Oilfield Xi'an: [Publisher unknown], 2019
13. [13] Du Hongyong, Sui Gu Lei Research on CNN based Fault Diagnosis Method for Pumping Unit Wells [J]. Contemporary Chemical Industry, 2021, 50 (12): 2909-2913
14. [14] Tang Shanjie Research on Productivity Prediction of Fractured Horizontal Wells in Tight Gas Reservoirs Based on Time Series Method [D]. Xi'an: Xi'an University of Petroleum, 2023
15. [15] Qi Shuxian, Yu Gaoming, Liu Yun, etc A new method for predicting oilfield recovery based on attention mechanism and long short-term memory network [J]. Contemporary Chemical Research, 2023 (9): 177-179
16. [16] Luo Faqiang, Liu Jingtao, Chen Xiuping, etc Intelligent prediction method for pore pressure in the 5th fault zone of Shunbei Oilfield based on BP and LSTM neural networks [J]. Petroleum Drilling and Production Technology, 2022, 44 (4): 506-514
17. [17] Chao Zheng, Feng Qihong, Zhang Xianmin, etc Prediction method for deep water plugging and profile control effect based on the combination of V/S and LSTM [C]//2021 International Conference on Oil and Gas Field Exploration and Development Qingdao: [Publisher unknown], 2021
18. [18] Li Jiaxing, Han Anxin, Pan Xiangshuai, etc Remaining oil time series prediction based on GAN-LSTM [C]//40th China Control Conference Shanghai: [Publisher unknown], 2021
19. [19] Li Yanhui, Wang Yanmeng Prediction and Effect of Oilfield Water Injection Flow Based on Data Enhancement Technology and CNN BiLSTM Attention [J]. Science, Technology and Engineering, 2023, 23 (32): 13896-13902
20. [20] Wang Jizhe Research and Application of Oil Well Production Prediction Algorithm Based on Deep Learning [D]. Xi'an: Xi'an University of Petroleum, 2022

Funding

National Natural Science Foundation of China Youth Science Foundation (52104020)

About the authors

Zhang Hanwen (1999-), male, Han ethnicity, from Weinan, Shaanxi. He is currently pursuing a master's degree in oil and gas field development engineering; Corresponding author: Xin Xiankang (1986-), male, Han ethnicity, from Wenshang, Shandong, associate professor, master's supervisor, PhD, research direction: oil and gas field development engineering