Data-driven deep-learning forecasting for oil production and pressure

被引:35
|
作者
Werneck, Rafael de Oliveira [1 ]
Prates, Raphael [1 ]
Moura, Renato [1 ]
Goncalves, Maiara Moreira [2 ]
Castro, Manuel [1 ]
Soriano-Vargas, Aurea [1 ]
Mendes Junior, Pedro Ribeiro [1 ]
Hossain, M. Manzur [2 ]
Zampieri, Marcelo Ferreira [2 ]
Ferreira, Alexandre [1 ]
Davolio, Alessandra [2 ]
Schiozer, Denis [2 ,3 ]
Rocha, Anderson [1 ]
机构
[1] Univ Estadual Campinas, Inst Comp, RECOD Ai, UNICAMP, BR-13083852 Campinas, SP, Brazil
[2] Univ Estadual Campinas, CEPETRO, UNICAMP, BR-13083970 Campinas, SP, Brazil
[3] Univ Estadual Campinas, Sch Mech Engn, BR-13083970 Campinas, SP, Brazil
关键词
Forecasting; Data-driven; Deep learning; Oil production; Pre-salt; NEURAL-NETWORK; PREDICTION;
D O I
10.1016/j.petrol.2021.109937
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Production forecasting plays an important role in oil and gas production, aiding engineers to perform field management. However, this can be challenging for complex reservoirs such as the highly heterogeneous carbonate reservoirs from Brazilian Pre-salt fields. We propose a new setup for forecasting multiple outputs using machine-learning algorithms and evaluate a set of deep-learning architectures suitable for time-series forecasting. The setup proposed is called N-th Day and it provides a coherent solution for the problem of forecasting multiple data points in which a sliding window mechanism guarantees there is no data leakage during training. We also devise four deep-learning architectures for forecasting, stacking the layers to focus on different timescales, and compare them with different existing off-the-shelf methods. The obtained results confirm that specific architectures, as those we propose, are crucial for oil and gas production forecasting. Although LSTM and GRU layers are designed to capture temporal sequences, the experiments also indicate that the investigated scenario of production forecasting requires additional and specific structures.
引用
收藏
页数:16
相关论文
共 50 条
  • [31] Deep spatial transformers for autoregressive data-driven forecasting of geophysical turbulence
    Chattopadhyay, Ashesh
    Mustafa, Mustafa
    Hassanzadeh, Pedram
    Kashinath, Karthik
    PROCEEDINGS OF 2020 10TH INTERNATIONAL CONFERENCE ON CLIMATE INFORMATICS (CI2020), 2020, : 106 - 112
  • [32] A Data-Driven Approach Using Deep Learning Time Series Prediction for Forecasting Power System Variables
    Kayedpour, Nezmin
    Samani, Arash E.
    De Kooning, Jeroen D. M.
    Vandevelde, Lieven
    Crevecoeur, Guillaume
    PROCEEDINGS OF 2019 IEEE 2ND INTERNATIONAL CONFERENCE ON RENEWABLE ENERGY AND POWER ENGINEERING (REPE 2019), 2019, : 43 - 47
  • [33] Data-driven models for forecasting algal biomass in a large and deep reservoir
    Li, Yuan
    Shi, Kun
    Zhu, Mengyuan
    Li, Huiyun
    Guo, Yulong
    Miao, Song
    Ou, Wei
    Zheng, Zhubin
    WATER RESEARCH, 2025, 270
  • [34] Model-driven deep-learning
    Zongben Xu
    Jian Sun
    National Science Review, 2018, 5 (01) : 22 - 24
  • [35] Deep-learning for seizure forecasting in canines with epilepsy
    Nejedly, Petr
    Kremen, Vaclav
    Sladky, Vladimir
    Nasseri, Mona
    Guragain, Hari
    Klimes, Petr
    Cimbalnik, Jan
    Varatharajah, Yogatheesan
    Brinkmann, Benjamin H.
    Worrell, Gregory A.
    JOURNAL OF NEURAL ENGINEERING, 2019, 16 (03)
  • [36] Physics-driven deep-learning for marine CSEM data inversion
    Liang, Hao
    Gao, Ruoyun
    Yin, Changchun
    Su, Yang
    He, Zhanxiang
    Liu, Yunhe
    JOURNAL OF APPLIED GEOPHYSICS, 2024, 229
  • [37] A Deep-Learning Framework for the Detection of Oil Spills from SAR Data
    Shaban, Mohamed
    Salim, Reem
    Abu Khalifeh, Hadil
    Khelifi, Adel
    Shalaby, Ahmed
    El-Mashad, Shady
    Mahmoud, Ali
    Ghazal, Mohammed
    El-Baz, Ayman
    SENSORS, 2021, 21 (07)
  • [38] A Hybrid Data-Driven Approach for Forecasting the Characteristics of Production Disruptions and Interruptions
    Bazargan-Lari, Mohammad Reza
    Taghipour, Sharareh
    INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING, 2022, 21 (04) : 1127 - 1154
  • [39] Data-Driven Photovoltaic Power Production Nowcasting and Forecasting for Polygeneration Microgrids
    Oneto, Luca
    Laureri, Federica
    Robba, Michela
    Delfino, Federico
    Anguita, Davide
    IEEE SYSTEMS JOURNAL, 2018, 12 (03): : 2842 - 2853
  • [40] Unraveling thermoelastic performance of the multi-directional functionally graded composite structure by data-driven deep-learning approach
    Wu, Xiaoling
    Chen, Qilong
    He, Wenhai
    Almadhor, Ahmad
    Abbas, Mohamed
    MECHANICS OF ADVANCED MATERIALS AND STRUCTURES, 2024, 31 (26) : 8465 - 8487