Data-driven model for hydraulic fracturing design optimization: focus on building digital database and production forecast

Abstract

Growing amount of hydraulic fracturing (HF) jobs in the recent two decades resulted in a significant amount of measured data available for development of predictive models via machine learning (ML). In multistage fractured completions, post-fracturing production analysis (e.g., from production logging tools) reveals evidence that different stages produce very non-uniformly, and up to 30% may not be producing at all due to a combination of geomechanics and fracturing design factors. Hence, there is a significant room for improvement of current design practices. We propose a data-driven model for fracturing design optimization, where the workflow is essentially split into two stages. As a result of the first stage, the present paper summarizes the efforts in the creation of a digital database of field data from several thousands of multistage HF jobs on vertical, inclined and near-horizontal wells from circa 20 different oilfields in Western Siberia, Russia. In terms of the number of points (fracturing jobs), the present database is a rare case of a representative dataset of about 5000 data points, compared to typical databases available in the literature, comprising tens or hundreds of points at best. Each point in the data base contains the vector of 92 input variables (the reservoir, well and the frac design parameters) and the vector of production data, which is characterized by 16 parameters, including the target, cumulative oil production. The focus is made on data gathering from various sources, data preprocessing and development of the architecture of the database as well as solving the production forecast problem via ML. Data preparation has been done using various ML techniques: the problem of missing values in the database is solved with collaborative filtering for data imputation; outliers are removed using visualization of cluster data structure by t-SNE algorithm. The production forecast problem is solved via CatBoost algorithm. Prediction capability of the model is measured with the coefficient of determination (R2) and reached 0.815. The inverse problem (selecting an optimum set of fracturing design parameters to maximize production) will be considered in the second part of the study to be published in another paper, along with a recommendation system for advising DESC and production stimulation engineers on an optimized fracturing design.