Fig 1.
Hydraulic-mechanical specific energy model and optimization flow diagram.
Table 1.
Drilling Data Table for the 455m - 465m Interval of Well Zhanghai A.
Table 2.
Drilling Data Table for the 3695m-3705m Interval of Well Zhanghai A.
Table 3.
Data Table for Imputing Missing Values of Weight on Bit Using Random Forest (455-465m).
Table 4.
Data Table for Imputing Missing Rotary Speed Values Using Random Forest (3695-3705m).
Table 5.
Data Table of Abnormal Weight on Bit for Well Zhanghai A (833-840m).
Table 6.
Data Table of Abnormal Weight on Bit for Well Zhanghai A (3095 - 3105m).
Fig 2.
a) Outlier Detection Performance of the 3σ Method; b) Outlier Detection Performance of the K-means Method; c)Outlier Detection Performance of the LOF Method (Black is the standard value, red is the abnormal value).
Table 7.
Data Table After S-G Filter Processing (833 - 840m).
Table 8.
Data Table After S-G Filter Processing (3095 - 3105m).
Table 10.
Drilling Parameters After Processing (Part of Well Zhanghai A).
Table 9.
On-site Drilling Parameters (Part of Well Zhanghai A).
Fig 3.
Comparison of Raw Data and Processed Data for Well Zhanghai A.
Fig 4.
Comparison of Raw Data and Processed Data from Well Zhanghai B.
Fig 5.
Comparison of Raw Data and Processed Data from Well Zhanghai C.
Fig 6.
Comparison of Raw Data and Processed Data from Well Zhanghai D.
Fig 7.
The RMSE comparison diagram of the original and processed data.
Table 11.
Summary of Fitting Parameters for the Ternary Model.
Fig 8.
Field Rate of Penetration (ROP) and the Corresponding Prediction Results of the Ternary Model; a) Well Zhanghai A; b)Well Zhanghai C; c)Well Zhanghai C; d)Well Zhanghai D.
Table 12.
Training Data for Well Zhanghai A (Partial).
Fig 9.
Regression Analysis of the Sample Set in the BP Neural Network.
Fig 10.
Prediction Performance of Rate of Penetration (ROP) for Well Zhanghai A.
Fig 11.
Prediction Performance of Rate of Penetration (ROP) for Well Zhanghai B.
Fig 12.
Prediction Performance of Rate of Penetration (ROP) for Well Zhanghai C.
Fig 13.
Prediction Performance of Rate of Penetration (ROP) for Well Zhanghai D.
Table 13.
Evaluation Indicators for the BP Neural Network Model.
Table 14.
Comparison of Prediction Accuracy.
Table 15.
Lithology Table for Part of the Strata in Well Zhanghai A.
Table 16.
Lithology Table for Part of the Strata in Well Zhanghai B.
Fig 14.
Mechanical Specific Energy Baseline; a) Well Zhanghai A; b) Well Zhanghai B.
Table 17.
Data Table for Correlation Analysis (Partial).
Fig 15.
Correlation Analysis of Parameters in Drilling Data.
Table 18.
Sensitivity Analysis of Weight on Bit (WOB) to Mechanical Specific Energy and Field Drilling Data.
Fig 16.
Variation Curve of Mechanical Specific Energy with Rate of Penetration (ROP) under Different Weight on Bit (WOB) Conditions.
Fig 17.
Variation Curve of Mechanical Specific Energy with Rate of Penetration (ROP) under Different Weight on Bit (WOB) Conditions.
Table 19.
Sensitivity Analysis of Rotational Speed on Mechanical Specific Energy.
Fig 18.
Variation Curve of Mechanical Specific Energy with Rate of Penetration (ROP) at Different Rotational Speeds.
Table 20.
Sensitivity Analysis of Flow Rate on Mechanical Specific Energy.
Fig 19.
Curves Depicting the Relationship Between Mechanical Specific Energy and Rate of Penetration (ROP) at Different Flow Rates.
Fig 20.
Corresponding Curve of Rate of Penetration (ROP) versus Mechanical Specific Energy (MSE) in the Case of Bit Balling.
Table 21.
Partial Parameter Values for the Mechanical Specific Energy Model.
Table 22.
Predicted Values of Rate of Penetration (ROP) for Well Zhanghai A (Partial).
Fig 21.
Limit Values of Parameters for Well Zhanghai A.
Table 23.
Parameter Constraint Conditions.
Fig 22.
Optimization Results of Drilling Parameters.
Table 24.
Lithology Table of Well Zhanghai A-1 Formation.
Fig 23.
Illustration of the Sliding Window Concept.
Table 25.
Sliding window quantitative data.
Fig 24.
Application Effect Diagram of Adjacent Wells.
Fig 25.
Comparison Chart of Transmission Efficiency.