Oil and Gas Field Big-Data Analytics

Oct 9, 2026 · 3 min read

Research / Area 6

Wellhead pressure, rate and chemical-injection data accumulate continuously as gas fields mature, but field decisions still rely largely on rules of thumb. The group uses data-driven modeling, risk-aware decision-making and deep anomaly diagnosis to move from post-hoc review to online recommendation.

Overall big-data analytics framework
Overall framework: field data → sample library → world model → planning agent → recommended dosage, in closed loop
Action-conditioned world model
Action-conditioned world model for next-step tubing–casing pressure-difference prediction
Risk-aware planning agent
Risk-aware planning agent (RAPA) with layered risk rewards
Dynamic injection actions comparison
Dynamic foaming-agent injection actions: RAPA vs. SAC, PSO, XGBoost and field baseline

Foam-assisted deliquification. With Prof. Jianli Wang’s group, foaming-agent dosage optimization is reformulated as a short-term decision problem driven by tubing–casing pressure-difference risk. An action-conditioned world model predicts the next pressure difference for candidate dosages, and a risk-aware planning agent (RAPA) selects actions with layered risk rewards.

Representative papers

  1. Y Zhang, H Wang, M Chen, N Li, J Zhu, J Wang. A data-driven world-model-based planning agent for risk-aware foam-lifting agent injection optimization in gas wells. Geoenergy Science and Engineering, manuscript submitted (GEOEN-S-26-02523), 2026.
  2. J Zhu, Y Liu, H Wang, M Chen, N Li, G Cao, R Zhong, H Zhu. Enhancing Plunger Lift Anomaly Detection: A Vision Transformer-Based Approach Leveraging Pretrained Models and Graphic Data Augmentation. Processes 14 (13), 2045, 2026.
  3. J Zhu, Y Liu, M Chen, H Wang, Y Li, H Zhu. Enhanced Zero-Shot Classification of Plunger Lift Operating Conditions Using a Modified Clip Architecture with Selective Data Sampling. Results in Engineering, 110368, 2026.
  4. Q Wang, H Zou, C Liang, S Zhang, H Jia, J Zhu. A combined approach for predicting liquid loading onset via clustering and convolutional neural networks. The Canadian Journal of Chemical Engineering, 2026.
  5. QX Liu, JJ Zhu, HB Wang, S Chen, HY Wang, N Li, RZ Zhong, YJ Liu et al. Deep Feature Learning for Anomaly Detection in Gas Well Deliquification using Plunger Lift: A Novel CNN-based Approach. Petroleum Science, 2025.
  6. L Peng, L Feng, Q Guang, J Zhu, H Liu, Z Nie, C Ma, C Di, Q Wu. Real-Time ESP Management Framework Using Hybrid Physics-Based and ML Models. SPE Offshore Europe Conference and Exhibition, D021S009R008, 2025.
  7. Z Zhong, H Wang, N Li, H Zhu, J Zhu, J Wang. Prediction of Plunger Lift Dynamics Using a Bidirectional Long Short-Term Memory Neural Network with an Innovative Forecasting Strategy. 2024 IEEE 3rd International Conference on Electrical Engineering, Big Data and Algorithms (EEBDA), 2024.
  8. Z Xu, H Lin, Y Jia, J Du, J Mao, J Zhu, J Chen, F Li, N Li. Data Augmentation and Recognition for Plunger Lift Based on Variational Autoencoders. 2024 6th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP), 2024.
  9. Y Xie, S Ma, H Wang, N Li, J Zhu, J Wang. Unsupervised clustering for the anomaly diagnosis of plunger lift operations. Geoenergy Science and Engineering 231, 212305, 2023.
  10. J Zhu, G Cao, W Tian, Q Zhao, H Zhu, J Song, J Peng, Z Lin, HQ Zhang. Improved data mining for production diagnosis of gas wells with plunger lift through dynamic simulations. SPE Annual Technical Conference and Exhibition, D021S030R001, 2019.