Enhancing Production Forecasting through Data Analytics and Machine Learning

The oil and gas industry is currently at a critical juncture, facing multifaceted challenges. To address these challenges, unconventional reservoirs, such as tight and shale formations, have gained prominence. Leveraging the potential of the Fourth Industrial Revolution, which incorporates digital technologies like AI, and big data analytics, the industry is poised to revolutionize the management of these reservoirs. This technical talk focuses on the pivotal role of data analytics and machine learning in advancing unconventional reservoir management. Two key aspects take center stage:

1. Outlier Detection: Traditional outlier detection methods have been found wanting, both in terms of efficacy and potential bias in production data time-series. The first part of the talk delves into the application of techniques of statistical, regression-based and machine learning methods for outlier detection to synthetic and real production data. Their advantages and performance are reviewed and discussed.

2. Advanced Machine Learning for Production Forecasting: The second facet explores innovative approaches to long-term production forecasting, a fundamental aspect of petroleum engineering. Probability Density Function-based Decline Curve Analysis models as well as advanced machine learning techniques including Long Short-Term Memory networks and Convolutional Neural Networks are introduced. These technologies empower reservoir and production engineers to capture intricate temporal and spatial nuances, even when historical data is short, resulting in more reliable and accurate long-term production forecasts.

In summary, this talk offers a holistic and technical solution to the two challenges associated with unconventional reservoirs. By enhancing data quality and accuracy through outlier detection and advancing production forecasting with sophisticated machine learning models, the proposed integrated approach paves the way for data-driven decision-making.

In-person registration closes on December 5 at 8AM MST.

CLICK HERE to REGISTER