Abstract:
In order to accurately predict the WOB, improve drilling efficiency and reduce drilling cost, a hybrid model combining Bidirectional Long Short-Term Memory Network (Bi-LSTM) and Convolutional Neural Network (CNN) was established. The model uses trigonometric function driven Particle Swarm Optimization (TDCSO) method for hyperparameter optimization to achieve high accuracy of WOB prediction. In the experiment, the public data sets of two Wells in FORGE 58-32 and FORGE 58-62 in Utah were used to verify the established model, and the model performance was evaluated by the mean absolute error, root mean square error, coefficient of determination and mean square error. The results show that the proposed TDCSO-CNN-Bi-LSTM model achieves excellent results in three key indicators of MAE, MSE and RMSE, and the
R^2 is as high as 0.98, which is significantly better than the existing LSTM, GRU, CNN-LSTM, CNN-Bi-LSTM and other methods. The study shows that the proposed TDCSO-CNN-Bi-LSTM model has excellent accuracy in DWOB prediction, enables real-time monitoring, and is integrated with an automated drilling system to achieve precise control of WOB. This not only improves the drilling efficiency, but also reduces the drilling cost, which has important practical application value for future drilling operations.