The world of energy is changing fast, driven by rising electricity demand, the push for decarbonisation, and the challenge of adding renewables to the grid.
In an exclusive interview with AIM, Shantanu Som, Asia executive engineering leader at GE Vernova, explained how cutting-edge tools could transform the way we keep power systems running reliably and efficiently.
What is Predictive Maintenance?
Predictive maintenance streamlines the maintenance process and involves prioritising tasks based on actual needs rather than a schedule. It helps allocate resources more efficiently and use them only when and where required.
This can also be implemented in the production of electricity, which is facing unprecedented demand due to the rapid growth of data centres, EV adoption, and the electrification of underserved regions.
With over 700 million people still without access to electricity, the world currently faces a huge challenge.
While phasing out coal demands massive investments, the rapid rise of renewable energy like solar and wind is straining power grids, causing supply instability. This is where predictive maintenance comes in.
It analyses the demand and supply chains for electricity to optimise and help keep things running smoothly, ensuring reliable energy while also working towards a cleaner, more sustainable future.
AI Solving Problems with Parity
GE Vernova has been using AI and machine learning for years with tools like Asset Performance Management (APM) systems and digital twins to keep the equipment running smoothly. APM uses sensor data to spot problems early on, while digital twins create virtual models of assets like gas turbines to compare real-time performance and detect issues.
“It’s like your car alerting you that your tyre will go flat in 5,000 kilometres, so you can act before that happens,” Shantanu explained. This approach helps avoid expensive downtime. Right now, these tools rely on physics models and past data.
Generative AI could take things further by solving complex problems and ensuring consistent results. With 7,000 gas turbines worldwide, it could share solutions across similar machines, thus cutting down on errors.
Shantanu further said that generative AI can help bring parity, improve maintenance and build trust in AI-driven systems.
A Cautious Approach to GenAI
Generative AI has great potential, yet its adoption in predictive maintenance is moving slowly. This is because, in industries like power and aviation, where safety and reliability are critical, results need to be precise and predictable.
“The challenge is we still don’t fully understand what generative AI can and cannot do,” said Shantanu. “In industries like ours, one must be sure about the outcome versus the randomness of the outcome. It’s the same reason aviation is cautious about new technologies.”
Salil Parekh, Infosys CEO and MD, said in July, “We are not at this stage disclosing and quantifying externally our revenue from it. The work we are doing is quite incredible. The focus is really on what enterprises are doing for generative AI.”
Randomness in generative AI poses a challenge that must be carefully managed. A cautious approach allows more time to test and refine the technology, ensuring it’s reliable and easy to understand before being deployed widely.
Diagnostics, as well as Troubleshooting
Generative AI may still be in its early days, but Shantanu sees a sea of potential. It could improve diagnostics by simulating complex scenarios and suggesting fixes before the problems occur. This would be especially useful as renewable energy systems get more complicated.
Moreover, it can create consistent solutions by analysing data from many machines and tailoring recommendations, reducing inconsistencies across fleets. Beyond maintenance, it could help design better, more efficient systems using past data and predictions.
Generative AI could also make troubleshooting faster by offering smart suggestions to quickly identify and fix issues, cutting downtime.
A Future Powered by AI
Predictive maintenance is just one piece of the puzzle. GE Vernova’s broader vision involves creating solutions that integrate renewables, gas power, and digital tools into a cohesive framework.
Platforms like the grid operating system are pivotal in managing grid stability by orchestrating which assets come online, how much power they generate, and for how long—all while keeping carbon intensity in check.
Generative AI could help manage grid challenges by simulating scenarios and finding the best ways to balance renewable energy fluctuations with reliable power supply.
While the industry is cautious, given the high stakes, the potential benefits are huge—more reliable systems, smoother operations, and lower carbon emissions. Generative AI’s role in the power sector reflects the energy transition itself: careful, gradual, and ultimately transformative.
“The acceptance will be slower. But I also feel that it has immense power, which will make decision-making much faster and curated right, which is not there today,” said Shantanu.
The post AI Powers Up to Tackle the World’s Toughest Energy Challenges appeared first on Analytics India Magazine.
