Real‐time prediction of horizontal drilling pressure based on convolutional Transformer.

Summary: During horizontal drilling operations, real‐time prediction of drilling pressure during the drilling process can help the drilling team cope with the complex and changing working environment downhole, adjust the parameters of the drilling rig promptly, make correct decisions, reduce the probability of drilling accidents, and avoid affecting the duration and cost of the project. This study provides a method for real‐time prediction of the drilling pressure of horizontal drilling rigs. A deep learning model based on a convolutional Transformer is trained for accurate real‐time prediction by extracting real‐time operating data of the horizontal drilling rig from the data acquisition system. The method proposed in this study can be a useful tool to improve the performance of horizontal drilling rigs and can assist the drilling team in operating horizontal drilling rigs. The results of the case study show that: (1) the proposed convolutional Transformer model provides reliable real‐time prediction with an MAE of 0.304 MPa and an RMSE of 0.508 MPa; (2) the proposed method can quickly and accurately predict the trend of drilling pressure change in the next period based on the current change of drilling pressure, and grasp the dynamics of drilling pressure of horizontal drilling rigs in advance. Further research could focus on assisted decision‐making and intelligent optimization to provide solutions for preventing drilling accidents and improving horizontal rig performance based on the prediction. [ABSTRACT FROM AUTHOR]