<img height="1" width="1" style="display:none;" alt="" src="https://px.ads.linkedin.com/collect/?pid=1719028&amp;fmt=gif">

How Aker BP is taking a hybrid
machine learning approach to optimizing production

Machine learning on its own is not a silver bullet for heavy-asset companies; however, the combination of machine learning and physics can deliver real value at scale.

Read how Aker BP saves an estimated $6 million a year through more efficient water contamination detection. 

In short

Aker BP is working on combining machine learning with physics and domain knowledge to monitor and improve produced water quality and reduce production losses.

How Aker BP is taking-01
$6 million
Estimated annual savings


Produced water disposal is one of many challenges at oil and gas facilities with high water-cut wells. Keeping the oil contamination level in the produced water below environmental limits requires an efficient separation process. This process is governed by a series of complex physical interactions. Significant production losses are associated with situations with high oil-in-water levels because safely discharging water to the sea requires slowing down production while troubleshooting for worst actors on the facility.



Aker BP, Expert Analytics, and Cognite have implemented a smart monitoring system that visualizes all data relevant for troubleshooting water contamination and is currently developing a recommender system with an underlying machine learning model to identify worst actors related to high oil-in-water concentrations.

The smart monitoring system displays near real-time data from Cognite Data Fusion (CDF) and is visualized in an intuitive Grafana dashboard. Additionally, calculations combining sensor values and simulator outputs provide engineers with virtual sensors and physical properties they otherwise would not have readily available.

The recommender system under development is based on a machine learning model fed with data from CDF. The model, which is based on historical data from approximately 200 sensors from production wells and equipment relevant to produced water, is trained to predict oil-in-water concentrations based on historical data and determines an importance associated with each parameter or property in the production facility. This in turn can be used to identify the worst actors related to water contamination.

In addition to the 200 physical sensors values, the model is also fed approximately 100 virtual sensor values. One example is how the pressure and temperature sensors are converted to fluid properties by applying the laws of thermodynamics and the information of the fluid composition. The model also considers virtual sensor values obtained from multiphase flow simulators. The flow rates from the individual wells are collected from a multiphase flow simulator.


The smart monitoring system provides one place for all relevant data for troubleshooting issues related to water contamination, increasing situational awareness. This allows the users to take informed actions based on the available data and solve the problem faster. When the proposed recommender system is in place, it will be an integral part of the monitoring system that will highlight which parts of the production facilities that may be important for the current oil-in-water concentration.

This serves as a starting point for investigating the problem, and is a significantly faster approach for identifying mitigating actions related to water contamination. Aker BP estimates that decreasing the time spent finding mitigating actions has an annual revenue potential of $6 million. The system could also have a net positive environmental impact.

Request a demo to witness
Cognite Data Fusion in action


Similar use cases

11th December 2019
How Cognite Data Fusion (CDF) supports manufacturing plant maintenance workers
Read more
11th December 2019
How Cognite Data Fusion enabled Framo and Aker BP to transition to a new smart service contract
Read more
11th December 2019
How Aarbakke and Cognite are extending the lifetime of CNC machines
Read more