Recap of Novi’s AAPG Ace Talk

On May 22 Novi’s chief geophysicist Kiran Sathaye gave a technical presentation at AAPG ACE 2019 in San Antonio on leveraging AI to predict hydrocarbon recovery. Read the abstract and be sure to view the presentation slides here.  Thank you to everyone who attended the talk!

Authors

K. Sathaye, J. Ramey, J. Wan

Abstract

We combine production and completion data from 9,000 unconventional wells in the Williston Basin with 42,000 geophysical log files representing 10,000 unique wells to build a predictive model for hydrocarbon production. We predict hydrocarbon recovery at 30-day intervals up to 2 years after the start of production. This model contains dozens of input features and incorporates their nonlinear, multivariate effects on hydrocarbon production. The subsurface modeling and recovery predictions are implemented using the open source machine learning tools Scikit-learn and Tensorflow.

In developing the subsurface model for this machine learning approach, we also create predictions for important subsurface features that are not commonly logged due to cost and complexity, such as measurements of hydrocarbon chain length in productive formations of the basin. These measurements show nonlinear relationships both in space and with commonly logged properties, boosting the impact of a relatively sparse dataset. Additionally, we use this public subsurface dataset to identify subdivisions in the Bakken (lower, middle, upper), and train the model with subsurface features informed by previous geologic studies which identified these sections in core studies and well log interpretation.

With only production and completion data included, we see less than 15-25% aggregate error for various areas of the basin. We show successive improvements in this machine learning model by training it with a sequence of datasets: Production and completion data with limited well counts and full dataset, and a model with subsurface features added with both limited well count and full dataset. For all models, we use an 80%-20% training-test split to avoid overfitting the model.

View Our AAPG Presentation Now!

Latest News

glossary(don't touch)

What is {{mpg_keyword}}? | Novi Labs

Oil & Gas Glossary {{mpg_keyword}}​ Confused by multitude of oil & gas data and analytics terms? Get easy-to-understand definitions in our glossary. Other terms What
Read More →
Release note

Novi Labs Releases Novi Data Network (NDN), Leveraging Proprietary Oil and Gas Data for AI Driven Well Forecasting Models

We are pleased to announce the release of Novi Data Network (NDN). NDN allows oil & gas operators to collectively pool their production and completion
Read More →
Exhibition / Event

Novi Attends URTeC 2023!

Novi is excited to be exhibiting at BOOTH 1341. and presenting seven papers! Click through for the full schedule.
Read More →

News, trends and data for the US upstream industry

Novi Energy Newsletter

Sign up today for free and get a weekly email with news and insights on important developments in one or two of the major tight oil & gas basins.

Name(Required)
This field is for validation purposes and should be left unchanged.

INTRODUCING CAUSAL MODELS

Accurate forecast on parent-child developments

In this live webinar, you will learn how Novi’s new algorithm improves model sensitivity for spacing and parent-child scenarios, providing powerful results for previously difficult-to-analyze problems.

Ted Cross, our VP of Product Management, will show you how this update improves spacing and infill scenario analysis without sacrificing model accuracy.