You are Home   »   News   »   View Article

Cognite's successful use cases

Wednesday, April 28, 2021

Cognite successfully built tools to help companies manage contaminated water, maintenance programs, turbine shutdowns, ship performance and risk factors.

Cognite has reported a number of successful 'use cases' - projects it developed for oil and gas and service companies, gathering, contextualising and visualising data to help its clients get real business results, in one case helping a client save $6m a year.

Here are summaries from some published use cases. You can read the full reports, some with screenshots, on

Contaminated water

Cognite worked with Aker BP to build a system to quickly detect contaminated water in oil production systems, helping staff to draw conclusions about which wells were likely to 'produce' contaminated water.

The project is estimated by Aker BP to save them $6m a year, by helping the company spot problems and react to them more quickly.

Wells often produce water mixed together with oil. The water needs to be separated from oil before it can be disposed of. This can be done using separation systems on offshore platforms, but if the amount of oil in the water is high, it may be better to throttle the well which is producing the water, or find some other way to reduce the contamination level.

The model built by Cognite uses data from 200 physical sensors on production wells and equipment relevant to produced water.

It also calculates 100 'virtual sensors', for example about fluid properties, which are inferred from pressure and temperature data, using thermodynamic calculations and information about the fluid composition.

The aim is to predict the oil in water concentration from each well, although it cannot be directly measured.

You can see near real time data on a dashboard, made using Grafana, an open source analytics and monitoring solution.

This dashboard can show engineers which part of their production facility might be the biggest drivers to their current oil-in-water concentration.

An Oslo data analytics consultancy called Expert Analytics was also involved in developing the data visualisation and advisory system.

Gas turbine monitoring

Cognite worked with oil and gas operator Wintershall DEA to build a system to monitor if a gas turbine was shut down properly.

The gas turbine is on the Mittelplate oilfield, generating electricity for offshore operations.

Cognite built a visualisation system for engineers showing them the health of the system in different phases of the shutdown, such as 'shutdown start', 'roll-out' and 'end'. They can see if any sensor data indicates a divergence from what is normally seen.

The data outputs are viewed on a Microsoft Power BI dashboard.

Before this was built, engineers were relying on alarms to say that something had passed a certain threshold, but they did not have any way to get early warnings about failures.

Wintershall DEA estimated that the solution saves them $865,000 per incident, due to problems with the shutdown.

Maintenance decision making

Cognite did a project for Wintershall DEA to gather a wide range of different maintenance data onto a dashboard.

It shows data about corrosion, component failure, and spending. Engineers can understand corrosion trends, see which components are least reliable, and see how well their predictive maintenance projets are working, compared to the costs of corrective maintenance.

It can cover the company's oil and gas production across multiple installations, pipelines, vehicles and other assets.

Monitoring pump health

Cognite put together a data model for pump manufacturer Framo, to monitor the condition of its seawater lift pumps on Aker BP's Ivar Aasen field.

As a result of the data model, Framo felt confident signing a contract with Aker BP where the compensation was linked to the uptime of the pumps, in a 6 year contract.

The pump monitoring data model is directly connected to systems for creating work orders for maintenance, and improving future product development.

Before the system was developed, Framo was not able to access much data about its equipment when it was in operation, unless there was a specific problem such as an equipment failure.

'With the release of these data flows, Framo has been able to predict the condition of equipment, foresee what will happen with the pumps in the future, and in turn plan effective maintenance,' said Trond Petter Abrahamsen, managing director of Framo Services AS.

Maintenance planning

Cognite built a system for Aker BP to be used for planning maintenance on critical equipment. It combines data from four different source systems onto a single dashboard, built using Microsoft Power BI.

The source data can include lube oil charts, time series sensor data, and maintenance work orders.

It has colour-coded labelling to help the offshore and onshore maintenance organisation decide where to focus maintenance resources.

The dashboard allows for deep dives into individual work orders to find relevant information such as estimated time to completion.

Before the system was built, Aker BP staff were manually compiling data onto a spreadsheet, which takes 15 hours to make. The spreadsheet was compiled before the company's weekly meeting to discuss priorities for maintenance of critical equipment. Now this is all done automatically.

Cognite and PGS

In February 2020, Cognite announced a digitalisation project with seismic imaging company PGS, to help find ways to improve the vessel's shipping fleet performance, energy consumption and maintenance.

It will actively monitor and contextualise data to optimise fuel consumption and work out ways to reduce maintenance costs.

Risk factors

Cognite developed a visualisation tool for risk data for Aker BP, combining data from multiple sources into one visualisation. It includes data on work permits, leaks, barrier impairments, well integrity status, noise levels and other temporary deviations.

The tool can be used to plan maintenance work, checking that there are no conflicts, such as high noise levels in areas work is planned to take place, or conflict between one planned task and another.

It is easy to add new information into the tool, such as a report about a strange noise.

It replaces a large amount of manual work which was done by Aker BP staff ahead of their daily work planning and risk assessment meeting compiling this data together.

Associated Companies
» Cognite
comments powered by Disqus


To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.


Latest Edition Jan Feb 2021
Feb 2021

Download latest and back issues


Learn more about supporting Digital Energy Journal