Our goal is less than 5 % error with 90 % less effort.
The main benefit of data-driven VFM is that it is effortless for your team
Data-driven methods deliver great superiority in terms of scalability and maintainability. That means that we can deliver live, quality assured rates with high accuracy at a competitive cost - with little to no effort required by your team. Combining this offering with our well test application will truly streamline your process from well tests to live rates.
Setting live rates up for existing wells, back-calculating your history or adding new wells are all effortless with data-driven methods. Ready to get a handle on your rates? Get in touch for a demo.
The magic is in cross-well learning
Our neural networks for estimating well rates are based on a unique architecture that allows the network to learn from the behavior of other wells. Data-driven methods are notoriously data hungry, so this innovation has allowed us to effectively multiply the amount of data our network can learn from.
This is particularly helpful for wells with little quality data, either because of lacking sensors and measurements, or because they are new with practically no history. Our database is growing continuously, benefiting all existing and future partners with a vast library to learn from.
Streamline by connecting uptream and downstream workflows
To reap the full benefits of live flow rates, they must be placed in a bigger context. The tasks that are upstream and downstream the well flow rate estimation problem and solution, must be included in the effort towards integration and automation. The data needed, of which well tests are the most important, must be systematically created, captured and mined. Automatic and online model training must be in place. Finally, model performance must be monitored continuously to inform better decision making. Our rate package module consists of a number of applications that streamline this process, and effortlessly links the rates to well testing and production optimization.