Full text loading...
-
Optimization of Multifractured Horizontal Well Performance Applying Machine Learning Techniques: Western Siberia Field Case Study
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, Saint Petersburg 2020, Nov 2020, Volume 2020, p.1 - 5
Abstract
Hydraulic fracturing is one of the most beneficial operations targeted to enhance oil production from unconventional reservoirs, and, certainly, the essential criterion of its success is properly planned hydraulic fracturing design. To make it optimal, the specialist should analyze plenty of appropriate data sources and decide which of them have the greatest impact on the outcome. It seems that machine learning algorithms are effective solution to the problem as they help finding hidden correlations between input and output variables (cumulative oil production, in this case) and highlight those which exert influence mostly. It is worth noting that one of the most valuable aspect of such approach is an opportunity to process vast amount of various data, which is directly relevant to the analysis from the engineer’s point of view. The goal of the research is to find the most robust algorithm able to forecast the target variable and define key hydraulic fracturing design parameters.