TY - CONF

T1 - Using Machine Learning Methods to Model and Interpret Problems in Determining the Parameters of the Medium Near the Well Using Signals from a High-Frequency Induction Logging While Drilling Device

AU - Vlasov, Alexander

AU - Al Masud, Rofikul

AU - Bykova, Galina

N1 - A. Vlasov, R. Al Masud and G. Bykova, "Using Machine Learning Methods to Model and Interpret Problems in Determining the Parameters of the Medium Near the Well Using Signals from a High-Frequency Induction Logging While Drilling Device," 2025 IEEE XVII International Scientific and Technical Conference on Actual Problems of Electronic Instrument Engineering (APEIE), Novosibirsk, Russian Federation, 2025, pp. 1-6, doi: 10.1109/APEIE66761.2025.11289409.

PY - 2025/11/14

Y1 - 2025/11/14

N2 - With the widespread use of horizontal directional drilling in the oil and gas industry, the need for accurate interpretation of surrounding formation properties has become critical. In horizontal wells, the entire wellbore surface is in contact with the reservoir, requiring drilling to occur in the most productive zones. High-Frequency Induction Logging While Drilling tools offer the ability to capture electromagnetic signals for evaluating near-wellbore formations. However, the nonlinear nature of signal response to geological parameters often makes interpretation difficult. Existing quantitative analysis systems are limited in scope, hard to integrate into real-time drilling workflows, and often rely on oversimplified assumptions. This paper introduces a machine learning-based approach to enhance the quantitative interpretation of signal data from high-frequency induction logging while drilling tools. The proposed models are capable of solving both forward and inverse problems, offering improved accuracy in signal prediction and formation property estimation. A custom Python-based software module was developed and tested using synthetic datasets designed to replicate geological conditions typical of Western Siberia. The results demonstrate high reliability in interpreting formation boundaries and signal responses, showing a clear advantage over conventional methods.

AB - With the widespread use of horizontal directional drilling in the oil and gas industry, the need for accurate interpretation of surrounding formation properties has become critical. In horizontal wells, the entire wellbore surface is in contact with the reservoir, requiring drilling to occur in the most productive zones. High-Frequency Induction Logging While Drilling tools offer the ability to capture electromagnetic signals for evaluating near-wellbore formations. However, the nonlinear nature of signal response to geological parameters often makes interpretation difficult. Existing quantitative analysis systems are limited in scope, hard to integrate into real-time drilling workflows, and often rely on oversimplified assumptions. This paper introduces a machine learning-based approach to enhance the quantitative interpretation of signal data from high-frequency induction logging while drilling tools. The proposed models are capable of solving both forward and inverse problems, offering improved accuracy in signal prediction and formation property estimation. A custom Python-based software module was developed and tested using synthetic datasets designed to replicate geological conditions typical of Western Siberia. The results demonstrate high reliability in interpreting formation boundaries and signal responses, showing a clear advantage over conventional methods.

KW - электромагнитное зондирование

KW - прибор ВИКПБ

KW - количественная интерпретация

KW - высокочастотный индукционный каротаж в процессе бурения

KW - горизонтальное бурение

KW - прямые и обратные задачи

KW - оценка свойств пласта

KW - тонкослоистые коллекторы

KW - характеристика недр

KW - Electromagnetic sounding

KW - VIKPB tool

KW - quantitative interpretation

KW - high-frequency induction logging while drilling

KW - horizontal drilling

KW - orward and inverse problems

KW - formation evaluation

KW - thinly layered reservoirs

KW - subsurface

KW - characterization

UR - https://www.scopus.com/pages/publications/105031783779

U2 - 10.1109/APEIE66761.2025.11289409

DO - 10.1109/APEIE66761.2025.11289409

M3 - Paper

SP - 1

EP - 6

T2 - 2025 IEEE XVII International Scientific and Technical Conference on Actual Problems of Electronic Instrument Engineering (APEIE)

Y2 - 14 November 2025 through 16 November 2025

ER -