Hear from the experts: How AI can help power and energy organizations

It’s no secret that power and energy organizations face monumental challenges:

Artificial intelligence (AI) and machine learning (ML) are not new to energy operations. To address these challenges, energy companies have adopted practices, using insights from software built with AI/ML to forecast equipment failures and optimize maintenance schedules. This enhances equipment availability while minimizing maintenance costs — helping ease the pressure on power generators.

Expanded use of AI in predictive maintenance has become a hotter topic, particularly with the rise of interest in agentic and generative AI, otherwise known as GenAI. However, energy companies need a responsible approach to choosing which AI technologies best address their realities and needs for reliable, secure, and safe production. Agentic and GenAI solutions are still evolving, and do not consistently provide accurate recommendations.

To assess the practical applications of AI in the power generation industry, POWER Magazine interviewed two experts from GE Vernova — Janet Webb (Program Management Director, Advanced Reliability), and Mazen Younes (Sr. Director, Platform & Essentials).

The full interview on using AI & predictive analytics to enhance power operations is available as a webinar.

In this article, we answer some of the key questions that are most often asked by GE Vernova customers.

The short answer is: Yes.

In the webinar, Webb uses a real customer example to showcase the impact that AI/ML can have on critical equipment.

The customer was using a GE Vernova HA gas turbine. Using GE Vernova’s APM SmartSignal, an equipment downtime prevention software, an assigned engineer received an advanced warning from the AI/ML SmartSignal digital twin, which detected an abnormal thrust bearing temperature increase.

From there:

Without AI/ML predictive analytics, this anomaly would have been missed — it wasn’t on the maintenance plan. Not only was a 6+ month unexpected outage prevented, but while waiting for the scheduled outage, parts were ordered and onsite in time for the planned outage.

Find more examples with AI/ML predictive analytics customer catches.

The GE Vernova experts explained that ML and deep learning are proven and mature AI technologies, which have been used by GE Vernova customers for more than a decade.

Younes shares how GE Vernova’s solutions, BoilerOpt and Autonomous Tuning – gas turbine performance software - use closed-loop neural network ML for combustion optimization. Customers using these solutions have significantly reduced their emissions and fuel costs by automatically optimizing for changes in temperatures, fuels, and equipment degradation.

A newer addition to GE Vernova’s product line is Autonomous Inspections – an automated visual inspection software - which detects issues using cameras and neural networks, providing insights to reduce manual errors and the time required for inspections .

So yes, AI can prevent surprises and failures. While GenAI may be getting the attention, mature AI technologies like machine learning and neural networks are already creating results for the power generation industry. Plus, many companies are still developing their parameters around using GenAI. In a survey during the webinar, about one-third of attendees said their company does not allow use of GenAI.

We’ve covered how AI can help reduce emissions and prevent surprise failures, but there are many other ways AI/ML can support power and energy customers.

Here are the top four ways AI/ML have been used by GE Vernova customers over the last 25 years:

An AI model is only as good as the data going into it. So, starting with good data is the first step to good results.

Yet, Webb explains that we often hear companies say they have the data but don’t know what to do with it. GE Vernova’s Asset Performance Management software helps these customers organize data so it’s centralized and accessible and provides insights that can be used to prevent failures, improve heat rate, or reduce emissions. With SmartSignal, for example, the data is presented in a way to empower plant teams to have visibility into what is running smoothly, where an issue is developing, and what needs attention. This helps take the guesswork out of maintenance prioritization.

About 30% of the webinar attendees said the benefit of adopting AI is not clear. Webb shares that if you feel overwhelmed with where to start or concerned about being ready for AI, GE Vernova has experts on hand to help you navigate that journey. You can start at one site and leverage our services and monitoring team to help you build a model with as little as two weeks of data. We also hear from customers who say they don’t think they’re ready for AI, but we offer flexible solutions and services to manage adoption at a rate they’re comfortable with. Often, it makes sense to start with one site or solution, realize the value, and grow from there.

It's normal to have this concern. However, Younes explains how GE Vernova thinks about it differently. It's about improving day-to-day operations, not replacing humans.

Whenever GE Vernova designs new AI tools, having a human in the loop is at the forefront of our workflow, keeping humans in control of outputs. We operate in a highly regulated industry and require human input and subject matter experts. We also recognize that some companies may not allow for Generative AI, so our solutions have opt-in and opt-out capabilities.

Even our closed-loop solutions are not intended to take a job, but to improve a role. Autonomous Tuning, for example, doesn’t replace an operator; they’re still required to control, operate, and maintain machines to meet power commitments. Automated gas turbine tuning finds the best combustion possible for a turbine every few seconds, enabling operators to focus on higher-level or more critical issues, while saving the company money and helping to meet emission regulatory requirements.

The ROI on asset performance management (APM) software for power generation can be quite significant.

Younes explains that if AI is not adopted, power companies face several challenges compared to those that do adopt AI:

AI solutions help prevent availability disruptions, which can be costly. Additionally, companies that don't optimize their energy generation may face higher energy costs. For example, without AI, companies may struggle to switch between renewable and gas energy sources efficiently, missing out on the most sustainable options during peak hours. Unlocking these benefits with tools built by expert vendors can lead to significant improvements.

Younes also cautions that safeguarding data and preventing IP leakages are crucial measures. It's safer to adopt AI gradually rather than all at once. Slower decision-making, higher operational costs, and availability disruptions are common issues. We must acknowledge the highly regulated industries and legal constraints to ensure a safe and compliant adoption of AI. It's better to proceed slowly and correctly than to rush or not adopt AI at all .

GE Vernova continuously evolves its solutions to meet the needs of power generation companies. Our solutions, for example, now include advanced features like drag-and-drop analytics and prescriptive recommendations. Our flexibility in providing both on-premises and cloud solutions ensures that we can tailor our offerings to the specific needs of each company.

Webb points out that GE Vernova has been leveraging AI/ML and neural network technology for over 20 years. Our solutions have been rigorously vetted and continuously improved to deliver the outcomes our customers expect.

Looking ahead, we’re investing more in AI, including autonomous inspection and prescriptive recommendation capabilities. As Generative AI technology matures and its value becomes clearer, we will integrate it to enhance our offerings while mitigating risks.
Choosing GE Vernova means adopting the right AI technology at the right time, backed by decades of experience in predictive analytics and a commitment to continuous improvement.

ⁱ Deloitte, Power and Utilities Industry Outlook 2025, https://www2.deloitte.com/us/en/insights/industry/power-and-utilities/power-and-utilities-industry-outlook.html

ⁱⁱ Occupational Health and Safety Administration (OSHA). Using Leading Indicators to Improve Safety and Health Outcomes. https://www.osha.gov/leading-indicators