The company used this data to calculate emissions per equipment as well as the specific energy efficiency for each asset. In all, it has monitored about 85% of emissions from operations. And now that these models are built, TotalEnergies can apply them to hundreds of other equipment and processes. The models provide real-time data calculations, analysis and KPIs to help the team identify and prioritize emissions reduction opportunities. In one use case, in which the company set out to optimize power delivery configuration of a site, it reduced CO2 emissions by 15% annually.

"Most of our sites are using [AVEVA] PI System, so most of the data is already available, but not really used properly to monitor the emissions, and this is why we decided to use [the asset framework function of AVEVA PI System] to design the proper templates to help them to monitor efficiently the emissions. Now the information is available to the operations people."

-Pierre Bernadi, Technical Advisor, TotalEnergies

Reducing critical equipment failures and operating more efficiently

In addition to the monitoring of greenhouse gas emissions, TotalEnergies wanted to reduce unplanned equipment downtime to drive more efficient operations. A few years ago, the company saw an increase in mechanical failures in critical equipment for purposes such as power generation, water injection, and gas compression in its growing fleet. Among the 105 breakdowns recorded, it discovered that half of them could have been avoided with proper remote monitoring and early warning detection. In 2019, an accident caused production to be halted for six months, which resulted in a loss of 120,000 barrels a day and several tens of millions of dollars (USD) in costs.

To mitigate and prevent such losses, the company implemented its Remote Assistance Intervention and Diagnosis (RAID) system. The system gathered data from equipment sensors fed to AVEVA™ PI Server, then fed that data to headquarters to run more advanced analytics using AVEVA Predictive Analytics. There was no additional cost to get data from the sensors, as this data and its storage was already in place. AVEVA Predictive Analytics and AVEVA PI System seamlessly work together to allow users to generate more insights. Users can cleanse AVEVA PI System data, develop no-code, AI-driven predictive models, and use prescriptive guidance in analyzing results to identify potential asset failures before they occur.

TotalEnergies deployed RAID upstream on 320 shaft lines, delivering a total of 36,000 alerts, with 429 critical equipment catches. Downstream, teams acted on 210 critical alerts. They avoided a production shortfall in upstream production, almost 500,000 barrels saved. Also, TotalEnergies avoided a catastrophic equipment failure like the one that happened in 2019.

Beyond these quantitative gains, the company has optimized maintenance and enhanced asset management. It has fostered internal capabilities to build other monitoring tools, such as online gas turbine carbon emissions monitoring, centralized centrifugal compressor performance monitoring, and a power reserve monitoring tool. TotalEnergies plans on focusing on wind next, adapting upstream activities and knowledge to this new industry. It has already deployed a remote monitoring system for one turbine in Scotland, with plans for mass deployment to a full wind farm soon.