Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method

To solve the problems of the low energy efficiency and slow penetration rate of drilling, we took the geological data of adjacent wells, real-time logging data, and downhole engineering parameters as inputs; the mechanical specific energy and unit footage cost as multi-objective optimization functio...

Full description

Bibliographic Details
Main Authors: Jinbo Song, Jianlong Wang, Bingqing Li, Linlin Gan, Feifei Zhang, Xueying Wang, Qiong Wu
Format: Article
Language:English
Published: MDPI AG 2022-10-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/15/21/8030
_version_ 1797468453370593280
author Jinbo Song
Jianlong Wang
Bingqing Li
Linlin Gan
Feifei Zhang
Xueying Wang
Qiong Wu
author_facet Jinbo Song
Jianlong Wang
Bingqing Li
Linlin Gan
Feifei Zhang
Xueying Wang
Qiong Wu
author_sort Jinbo Song
collection DOAJ
description To solve the problems of the low energy efficiency and slow penetration rate of drilling, we took the geological data of adjacent wells, real-time logging data, and downhole engineering parameters as inputs; the mechanical specific energy and unit footage cost as multi-objective optimization functions; and the machine pump equipment limit as the constraint condition. A constrained Bayesian optimization algorithm model was established for the optimization solution, and drilling parameters such as weight-of-bit, revolutions per minute, and flowrate were optimized in real time. Through a comparison with NSGA-II, random search, and other optimization algorithms, and the application results of example wells, we show that the established Bayesian optimization algorithm has a good optimization effect while maintaining timeliness. It is suitable for real-time optimization of drilling parameters, can aid a driller in identifying the drilling rate and potential tapping area, and provides a decision-making basis for avoiding the low-efficiency rock-breaking working area and improving rock-breaking efficiency.
first_indexed 2024-03-09T19:07:41Z
format Article
id doaj.art-b0283b6e82f14c0498ba1b3a184a2ce9
institution Directory Open Access Journal
issn 1996-1073
language English
last_indexed 2024-03-09T19:07:41Z
publishDate 2022-10-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj.art-b0283b6e82f14c0498ba1b3a184a2ce92023-11-24T04:30:45ZengMDPI AGEnergies1996-10732022-10-011521803010.3390/en15218030Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian MethodJinbo Song0Jianlong Wang1Bingqing Li2Linlin Gan3Feifei Zhang4Xueying Wang5Qiong Wu6School of Petroleum Engineering, Yangtze University, Wuhan 430100, ChinaEngineering and Technology Research Institute of CNPC Bohai Drilling Engineering Co., Ltd., Tianjin 300457, ChinaCNPC Xibu Drilling Engineering Co., Ltd., Urumqi 830011, ChinaEngineering and Technology Research Institute of CNPC Bohai Drilling Engineering Co., Ltd., Tianjin 300457, ChinaSchool of Petroleum Engineering, Yangtze University, Wuhan 430100, ChinaSchool of Petroleum Engineering, Yangtze University, Wuhan 430100, ChinaPetroChina Tarim Oilfield Branch, Korla 841000, ChinaTo solve the problems of the low energy efficiency and slow penetration rate of drilling, we took the geological data of adjacent wells, real-time logging data, and downhole engineering parameters as inputs; the mechanical specific energy and unit footage cost as multi-objective optimization functions; and the machine pump equipment limit as the constraint condition. A constrained Bayesian optimization algorithm model was established for the optimization solution, and drilling parameters such as weight-of-bit, revolutions per minute, and flowrate were optimized in real time. Through a comparison with NSGA-II, random search, and other optimization algorithms, and the application results of example wells, we show that the established Bayesian optimization algorithm has a good optimization effect while maintaining timeliness. It is suitable for real-time optimization of drilling parameters, can aid a driller in identifying the drilling rate and potential tapping area, and provides a decision-making basis for avoiding the low-efficiency rock-breaking working area and improving rock-breaking efficiency.https://www.mdpi.com/1996-1073/15/21/8030drilling parameterreal-time optimizationmulti-objective optimization problemconstrained Bayesian algorithm
spellingShingle Jinbo Song
Jianlong Wang
Bingqing Li
Linlin Gan
Feifei Zhang
Xueying Wang
Qiong Wu
Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
Energies
drilling parameter
real-time optimization
multi-objective optimization problem
constrained Bayesian algorithm
title Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
title_full Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
title_fullStr Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
title_full_unstemmed Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
title_short Real-Time Drilling Parameter Optimization Model Based on the Constrained Bayesian Method
title_sort real time drilling parameter optimization model based on the constrained bayesian method
topic drilling parameter
real-time optimization
multi-objective optimization problem
constrained Bayesian algorithm
url https://www.mdpi.com/1996-1073/15/21/8030
work_keys_str_mv AT jinbosong realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT jianlongwang realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT bingqingli realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT linlingan realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT feifeizhang realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT xueyingwang realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod
AT qiongwu realtimedrillingparameteroptimizationmodelbasedontheconstrainedbayesianmethod