Machine Learning Clustering of Reservoir Heterogeneity with Petrophysical and Production Data

Konoshonkin, D.; Shishaev, G.; Matveev, Ivan; Volkova, Aleksandra; Rukavishnikov, Valeriy S.; Demyanov, Vasily; Belozerov, Boris

doi:10.2118/200614-ms

Cited by 8 publications

(1 citation statement)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Agito and Bjorkevoll [34] proposed a hybrid approach between machine learning (ML) and physics-based modeling to provide decision support for drilling problems using python scripting. Konoshonkin, et al [35] proposed a metric-based machine-learning approach to identify and describe spatial trends in reservoir heterogeneity/facies property distribution using wireline and production data. They demonstrated how the proposed method can help to partition reservoir heterogeneity and discover and verify spatial trends for a real mature producing field in Western Siberia.…”

Section: Machine Learning With Pythonmentioning

confidence: 99%

Review of Python Applications in Solving Oil and Gas Problems

Okon

Appah

Ajienka

2021

JERR

View full text Add to dashboard Cite

Python has grown in popularity throughout various industries, corporations, universities, government, and research groups. Its true potential to automate various processes while increased predictability capabilities have been noticed in various industries. The petroleum industry is at the beginning phase of applying it to solve oil and gas problems. The rise in its popularity in the oil and gas industry is due to the digital transformation such as sensors and high-performance computing services that enable artificial intelligence (AI), machine learning (ML), big data acquisition, and storage in digital oilfields. A quick search on the number of publications in the oil and gas industry with the Society of Petroleum Engineers (OnePetro) in the past few years attests to this fact. Hence, it has proven to be a promising application that can bring about a revolutionary change in the oil and gas industry and transform the existing features for solving oil and gas problems. This will help the production and reservoir engineers to better manage the production operation without any need for expensive software. It will also reduce the overall operating cost and increase revenue.

show abstract

Section: Machine Learning With Pythonmentioning

confidence: 99%

Review of Python Applications in Solving Oil and Gas Problems

Okon

Appah

Ajienka

2021

JERR

View full text Add to dashboard Cite

show abstract

A novel reservoir classification method for sandstone reservoir evaluation using multi-scale digital rock method

Chen,

Gao,

Wang

et al. 2024

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

The formulation of an effective development plan, optimizing exploitation efficiency, relies heavily on accurate reservoir characterization. However, existing reservoir classification methodologies, which rely on macroscopic geological observations and well-logging data, have limitations in data availability and fail to capture micro-scale pore structure characteristics accurately. The emerging technology of digital rock, widely employed in the petroleum industry, also faces challenges in accurately capturing reservoir heterogeneity due to its single-scale nature and limited application for reservoir classification. To overcome these challenges, this study presents an innovative reservoir classification approach based on the multi-scale digital rock method. This approach integrates qualitative assessment of micro-CT images for particle contact relationships and pore types with quantitative evaluation of multi-scale characteristic parameters, obtained from micro-CT and scanning electron microscopy (SEM) images. The proposed classification method demonstrates excellent agreement with the prioritization of development and production data for offshore sandstone reservoirs. By establishing a robust evaluation methodology, this classification method facilitates exploration potential assessment and development strategy optimization within the field.

show abstract

A Computational Model for Wells’ Performance Analysis

Ita

Appah

2021

Day 2 Tue, August 03, 2021

View full text Add to dashboard Cite

The ability to identify underperforming wells and recover the remaining oil in place is a cornerstone for effective reservoir management and field development strategies. As advancement in computing programming capabilities continuous to grow, Python has become an attractive method to build complicated statistical models that predicts, diagnose or analyze well performance, efficiently and accurately. The aim of this study is to develop a computational model that will allows us to diagnose and analyze well performance using nodal analysis with the help of python. In this study, python was used to compute Nodal analysis method using Darcy and Vogel Equations. A case study was carried out using the data obtained from a field operating in the Niger Delta. Again, sensitivity of tubing size was conducted using python. The results obtained showed that a computational model with python has the ability to visualize, model and analyze wells performances. This technique will petroleum engineers to better monitor evaluate and enhance their production operation without the need for expensive softwares. This will reduce operating cost increases revenue.

show abstract

Machine Learning Clustering of Reservoir Heterogeneity with Petrophysical and Production Data

Cited by 8 publications

References 1 publication

Review of Python Applications in Solving Oil and Gas Problems

Review of Python Applications in Solving Oil and Gas Problems

A novel reservoir classification method for sandstone reservoir evaluation using multi-scale digital rock method

A Computational Model for Wells’ Performance Analysis

Contact Info

Product

Resources

About