Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration
Prediction of the rate of penetration (ROP) is integral to drilling optimization. Many scholars have established intelligent prediction models of the ROP. However, these models face challenges in adapting to different formation properties across well sections or regions, limiting their applicability...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-07-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/16/15/5670 |
_version_ | 1797586843079802880 |
---|---|
author | Jianxin Ding Rui Zhang Xin Wen Xuesong Li Xianzhi Song Baodong Ma Dayu Li Liang Han |
author_facet | Jianxin Ding Rui Zhang Xin Wen Xuesong Li Xianzhi Song Baodong Ma Dayu Li Liang Han |
author_sort | Jianxin Ding |
collection | DOAJ |
description | Prediction of the rate of penetration (ROP) is integral to drilling optimization. Many scholars have established intelligent prediction models of the ROP. However, these models face challenges in adapting to different formation properties across well sections or regions, limiting their applicability. In this paper, we explore a novel prediction framework combining feature construction and incremental updating. The framework fine-tunes the model using a pre-trained ROP representation. Our method adopts genetic programming to construct interpretable features, which fuse bit properties with engineering and hydraulic parameters. The model is incrementally updated with constant data streams, enabling it to learn the static and dynamic data. We conduct ablation experiments to analyze the impact of interpretable features’ construction and incremental updating. The results on field drilling datasets demonstrate that the proposed model achieves robustness against forgetting while maintaining high accuracy in ROP prediction. The model effectively extracts information from data streams and constructs interpretable representational features, which influence the current ROP, with a mean absolute percentage error of 7.5% on the new dataset, 40% lower than the static-trained model. This work provides a theoretical reference for the interpretability and transferability of ROP intelligent prediction models. |
first_indexed | 2024-03-11T00:28:53Z |
format | Article |
id | doaj.art-ec51f33c68fb4ba0a13d40a4c92d952e |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-03-11T00:28:53Z |
publishDate | 2023-07-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-ec51f33c68fb4ba0a13d40a4c92d952e2023-11-18T22:51:19ZengMDPI AGEnergies1996-10732023-07-011615567010.3390/en16155670Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of PenetrationJianxin Ding0Rui Zhang1Xin Wen2Xuesong Li3Xianzhi Song4Baodong Ma5Dayu Li6Liang Han7Kunlun Digital Technology Co., Ltd., Beijing 100043, ChinaCollege of Artificial Intelligence, China University of Petroleum (Beijing), Beijing 102249, ChinaKunlun Digital Technology Co., Ltd., Beijing 100043, ChinaKunlun Digital Technology Co., Ltd., Beijing 100043, ChinaCollege of Artificial Intelligence, China University of Petroleum (Beijing), Beijing 102249, ChinaNational Key Laboratory of Petroleum Resources and Engineering, China University of Petroleum (Beijing), Beijing 102249, ChinaNational Key Laboratory of Petroleum Resources and Engineering, China University of Petroleum (Beijing), Beijing 102249, ChinaNational Key Laboratory of Petroleum Resources and Engineering, China University of Petroleum (Beijing), Beijing 102249, ChinaPrediction of the rate of penetration (ROP) is integral to drilling optimization. Many scholars have established intelligent prediction models of the ROP. However, these models face challenges in adapting to different formation properties across well sections or regions, limiting their applicability. In this paper, we explore a novel prediction framework combining feature construction and incremental updating. The framework fine-tunes the model using a pre-trained ROP representation. Our method adopts genetic programming to construct interpretable features, which fuse bit properties with engineering and hydraulic parameters. The model is incrementally updated with constant data streams, enabling it to learn the static and dynamic data. We conduct ablation experiments to analyze the impact of interpretable features’ construction and incremental updating. The results on field drilling datasets demonstrate that the proposed model achieves robustness against forgetting while maintaining high accuracy in ROP prediction. The model effectively extracts information from data streams and constructs interpretable representational features, which influence the current ROP, with a mean absolute percentage error of 7.5% on the new dataset, 40% lower than the static-trained model. This work provides a theoretical reference for the interpretability and transferability of ROP intelligent prediction models.https://www.mdpi.com/1996-1073/16/15/5670interpretable feature constructiongenetic programmingrate of penetrationincremental updatefine-tune |
spellingShingle | Jianxin Ding Rui Zhang Xin Wen Xuesong Li Xianzhi Song Baodong Ma Dayu Li Liang Han Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration Energies interpretable feature construction genetic programming rate of penetration incremental update fine-tune |
title | Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration |
title_full | Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration |
title_fullStr | Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration |
title_full_unstemmed | Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration |
title_short | Interpretable Feature Construction and Incremental Update Fine-Tuning Strategy for Prediction of Rate of Penetration |
title_sort | interpretable feature construction and incremental update fine tuning strategy for prediction of rate of penetration |
topic | interpretable feature construction genetic programming rate of penetration incremental update fine-tune |
url | https://www.mdpi.com/1996-1073/16/15/5670 |
work_keys_str_mv | AT jianxinding interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT ruizhang interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT xinwen interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT xuesongli interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT xianzhisong interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT baodongma interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT dayuli interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration AT lianghan interpretablefeatureconstructionandincrementalupdatefinetuningstrategyforpredictionofrateofpenetration |