Jun Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jun Liu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Bang Zhou
Tianjin University
4 shared publications - Hanqing Chen
Tianjin University
4 shared publications - Mengyu Li
Tianjin University
4 shared publications - Ruiqi Wang
Tianjin University
4 shared publications - Weidong Cao
SINOPEC Shengli Oilfield Company
4 shared publications - Zhongke Gao
Tianjin University
4 shared publications
Latest Publications
- 25 June 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 295
LAPNet: A lightweight adaptive patch network for measuring water cut in oil-water two-phase flow
- 25 June 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 295
LAPNet: A lightweight adaptive patch network for measuring water cut in oil-water two-phase flow
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
APMNet-Based Water-Cut Prediction Using Dual-Helical Microwave Sensing
In industrial settings of high water-cut oil-water two-phase flow, accurately measuring wellhead parameters is essential for effective production monitoring and optimization in oilfields. The complexity of oil-water flow interactions and the limitations of current techniques pose significant challenges. We design a dual-helical microwave sensor, optimized through simulations for its structure and frequency. By integrating microwave transmission line technology with an innovative dual-helical design, the sensor enhances measurement stability and accuracy. Dynamic experiments are conducted to capture fluid signals under various conditions. We then propose an adaptive patch multi-feature fusion network (APMNet) for signal fusion and analysis. The dynamic patch module adjusts patch positions and sizes based on sequence variations, capturing more local changes and richer features, while the multi-feature fusion module integrates the features across the time, patch, and channel dimensions, effectively addressing issues of missing local features and noise interference. This approach achieves high-precision water cut prediction with a mean absolute percentage error (MAPE) of 2.08%. Comparative and ablation studies demonstrate the superior performance of our method in parameter prediction. Furthermore, field deployment in active oilfields confirms the practical effectiveness of the system in real-world applications.
- 2 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-10
APMNet-Based Water-Cut Prediction Using Dual-Helical Microwave Sensing
In industrial settings of high water-cut oil-water two-phase flow, accurately measuring wellhead parameters is essential for effective production monitoring and optimization in oilfields. The complexity of oil-water flow interactions and the limitations of current techniques pose significant challenges. We design a dual-helical microwave sensor, optimized through simulations for its structure and frequency. By integrating microwave transmission line technology with an innovative dual-helical design, the sensor enhances measurement stability and accuracy. Dynamic experiments are conducted to capture fluid signals under various conditions. We then propose an adaptive patch multi-feature fusion network (APMNet) for signal fusion and analysis. The dynamic patch module adjusts patch positions and sizes based on sequence variations, capturing more local changes and richer features, while the multi-feature fusion module integrates the features across the time, patch, and channel dimensions, effectively addressing issues of missing local features and noise interference. This approach achieves high-precision water cut prediction with a mean absolute percentage error (MAPE) of 2.08%. Comparative and ablation studies demonstrate the superior performance of our method in parameter prediction. Furthermore, field deployment in active oilfields confirms the practical effectiveness of the system in real-world applications.