Application of machine learning in predicting oil rate decline for Bakken shale oil wells

Abstract

Commercial reservoir simulators are required to solve discretized mass-balance equations. When the reservoir becomes heterogeneous and complex, more grid blocks can be used, which requires detailed and accurate reservoir information, for e.g. porosity, permeability, and other parameters that are not always available in the field. Predicting the EUR (Estimated Ultimate Recovery) and rate decline for a single well can therefore take hours or days, making them computationally expensive and time-consuming. In contrast, decline curve models are a simpler and speedier option because they only require a few variables in the equation that can be easily gathered from the wells' current data. The well data for this study was gathered from the Montana Board of Oil and Gas Conservation's publicly accessible databases. The SEDM (Stretched Exponential Decline Model) decline curve equation variables specifically designed for unconventional reservoirs variables were correlated to the predictor parameters in a random oil field well data set. The study examined the relative influences of several well parameters. The study's novelty comes from developing an innovative machine learning (ML) (random forest (RF)) based model for fast rate-decline and EUR prediction in Bakken Shale oil wells. The successful application of this study relies highly on the availability of good quality and quantity of the dataset.

New Model for Predicting Production Capacity of Horizontal Well Volume Fracturing in Tight Reservoirs

“…Al-Sabaeei et al used random forest and principal component analysis to predict the change law of production capacity in tight reservoirs and further discuss the applicability of machine learning for recovery prediction . Bhattacharyya and Vyas proposed a new theory based on the ANN model, which can reduce the error by the weight and bias method and improve accuracy using the genetic algorithm and ion swarm method . Al-Mudhafar et al used the genetic algorithm to predict the changing pattern of the bottom-hole pressure during the production of horizontal wells.…”

“… 46 Bhattacharyya and Vyas proposed a new theory based on the ANN model, which can reduce the error by the weight and bias method and improve accuracy using the genetic algorithm and ion swarm method. 47 Al-Mudhafar et al used the genetic algorithm to predict the changing pattern of the bottom-hole pressure during the production of horizontal wells. The error level was reduced to 10% compared with the measured data.…”

Real-time prediction of multivariate ROP (rate of penetration) based on machine learning regression algorithms: Algorithm comparison, model evaluation and parameter analysis

“…Meanwhile, many scholars have also used machine learning algorithms for reservoir simulation studies in the field of oil and gas field development in recent years. Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c) developed a novel data-driven-based model that could accurately predict the decline curves and EUR (Estimated Ultimate Recovery) for new wells based on the data collected from nearby wells and this paper presented use of new algorithm as well as a new dataset. Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c) developed an innovative machine learning (ML) (random forest (RF)) based model for fast rate-decline and EUR prediction in Bakken Shale oil wells.…”

“…Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c) developed a novel data-driven-based model that could accurately predict the decline curves and EUR (Estimated Ultimate Recovery) for new wells based on the data collected from nearby wells and this paper presented use of new algorithm as well as a new dataset. Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c) developed an innovative machine learning (ML) (random forest (RF)) based model for fast rate-decline and EUR prediction in Bakken Shale oil wells. Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c presented a novel approach for reservoir simulation that used Random Forest (RF) which is one of the widely used Machine learning (ML) algorithm to reduce the number of iterations at each time step and speed up the simulation process.…”

“…Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c) developed an innovative machine learning (ML) (random forest (RF)) based model for fast rate-decline and EUR prediction in Bakken Shale oil wells. Bhattacharyya et al (2022aBhattacharyya et al ( , 2022bBhattacharyya et al ( , 2022c presented a novel approach for reservoir simulation that used Random Forest (RF) which is one of the widely used Machine learning (ML) algorithm to reduce the number of iterations at each time step and speed up the simulation process. The study's novelty was in developing a new ML-based reservoir simulator that would make reservoir simulation much faster and computationally more efficient.…”