Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells

Wu, Zebing; Song, Lianghui; Xu, Jun; Chen, Jian; Wang, Tianci; Wang, Qinglin; Wang, Siqi

doi:10.3390/s26154747

This is an early access version, the complete PDF, HTML, and XML versions will be available soon.

Open AccessArticle

Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells

by

Zebing Wu

^1,*

,

Lianghui Song

^1,*,

Jun Xu

²,

Jian Chen

²,

Tianci Wang

¹

,

Qinglin Wang

¹ and

Siqi Wang

¹

Mechanical Engineering College , Xi’an Petroleum University, Xi’an 710065, China

²

Sinopec SJ Petroleum Machinery Co., Jingzhou 434024, China

^*

Authors to whom correspondence should be addressed.

Sensors 2026, 26(15), 4747; https://doi.org/10.3390/s26154747

Submission received: 18 June 2026 / Revised: 15 July 2026 / Accepted: 23 July 2026 / Published: 26 July 2026

(This article belongs to the Section Industrial Sensors)

Download Versions Notes

Abstract

Accurate estimation of downhole weight on bit (DWOB) is important for drilling parameter optimization and safe drilling in extended-reach horizontal wells. However, DWOB is difficult to obtain continuously because surface weight on bit (SWOB) is attenuated by drill-string friction, wellbore trajectory, and borehole-wall contact. To address this problem, a physics-informed convolutional neural network combined with bidirectional gated recurrent unit (CNN-BiGRU) prediction model optimized by the whale optimization algorithm (WOA) is proposed using surface sensor data and downhole measurement-while-drilling (MWD) measurements. The model first uses a one-dimensional CNN to extract local features from drilling parameters along the measured-depth direction and then employs a BiGRU to capture depth-series dependencies. Meanwhile, a drill-string frictional attenuation relationship is embedded into the loss function as a physical prior, and WOA is used to optimize network hyperparameters and the physics-constrained weight. Field data from a horizontal well were used for validation. The proposed model achieved a coefficient of determination (R²)of 0.9245 on the test set, with root mean square error (RMSE), mean absolute error (MAE), and mean relative error (MRE) values of 0.1856, 0.1135, and 0.0124, respectively. The results demonstrate that the proposed data–physics hybrid framework improves the accuracy and physical consistency of DWOB prediction.

Keywords: horizontal well; DWOB; surface sensor data and downhole MWD measurements; physics-informed neural network (PINN); CNN-BiGRU; WOA; drill-string frictional attenuation

Share and Cite

MDPI and ACS Style

Wu, Z.; Song, L.; Xu, J.; Chen, J.; Wang, T.; Wang, Q.; Wang, S. Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells. Sensors 2026, 26, 4747. https://doi.org/10.3390/s26154747

AMA Style

Wu Z, Song L, Xu J, Chen J, Wang T, Wang Q, Wang S. Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells. Sensors. 2026; 26(15):4747. https://doi.org/10.3390/s26154747

Chicago/Turabian Style

Wu, Zebing, Lianghui Song, Jun Xu, Jian Chen, Tianci Wang, Qinglin Wang, and Siqi Wang. 2026. "Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells" Sensors 26, no. 15: 4747. https://doi.org/10.3390/s26154747

APA Style

Wu, Z., Song, L., Xu, J., Chen, J., Wang, T., Wang, Q., & Wang, S. (2026). Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells. Sensors, 26(15), 4747. https://doi.org/10.3390/s26154747

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Physics-Informed CNN-BiGRU Model for Downhole Weight-on-Bit Prediction Using Surface Measurement-While-Drilling Sensor Data in Horizontal Wells

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI