Decarbonisation Technology - February 2025 Issue

System (EOS), provides precise control and dynamically integrates energy technologies and assets into a cohesive network. Intelligent, real-time decision- making without the need for continuous human involvement is coupled with continuous learning. AI-driven EMS technologies create and continuously update a site’s load profile, allowing for strategic load-planning. Changes in energy usage can be detected and responded to in

Figure 2 Inside a battery energy storage system

power supply (UPS). For complex sites with high energy usage, a BESS with UPS is far more efficient than traditional UPS for emergency power. For one Powerstar client, switching from lead-acid battery technology to a modern BESS with UPS has reduced annual energy spend by approximately £225,000 and cut 190 tonnes of CO₂ emissions. The IEA estimates that more than half of global O&G production currently lies within 10km of an electricity grid, with three-quarters in areas with good wind or solar resources. While offshore sites will generally be more expensive to electrify, the IEA considers that grid connections can be a viable option for onshore O&G fields (IEA, 2024 p78) . Where a grid connection is feasible, the BESS can be used to draw down grid energy when at its lowest price and stored for use at peak times while maximising the use and flexibility of any on-site renewables, thus reducing Scope 2 emissions through effective and efficient management of grid energy alongside clean energy. Harnessing the power of AI for data management and energy efficiency For optimum management across a microgrid – where different assets, such as the BESS, on-site renewable generation, and electrified heat pumps are incorporated – AI-driven management controls are critical. These enable O&G companies to benefit from data collection and analysis for continual energy efficiency improvement. A neural network-based AI-enabled controller, an energy management system (EMS) such as Powerstar’s proprietary Energy Optimisation

real-time, and faults or power disruptions can be pinpointed. Power and resource demands can be adjusted for optimal performance across a site’s energy management infrastructure. Given the complexity of O&G infrastructure and the lack of a clear, financially profitable trajectory for decarbonisation, AI-led continuous learning can be an important asset for companies embarking on electrification. Sensors and equipment across a site can provide real-time data on energy usage, whereby the central computer actively learns and adapts, facilitating efficient site management as larger quantities of data become available. AI integration empowers the management system, allowing it to improve performance progressively and to adapt to changing conditions. Data generated by remote monitoring across the microgrid is accessible via a user- friendly interface, allowing for better- informed and better-documented decisions. All of this contributes to a proactive, ongoing energy transition strategy. De-risking complexity: How a digital twin can inform and justify electrification Every O&G company embarking on or extending electrification needs to justify investment and change. Advanced modelling and simulation – the deployment of digital twins – can inform a sustainability strategy, enabling the evaluation of a multitude of scenarios and establishing the case for investment through data-driven cost-benefit analysis (see Figure 3 ). Complex projects – the interplay of processes and a range of variables – can be tested to determine their efficiency and effectiveness.