It is already time to plan for next hurricane season

In 2024, U.S. electricity customers experienced an average of 11 hours of electricity interruptions—nearly twice the annual average over the preceding decade. Major events such as Hurricanes accounted for 80% of the hours without electricity.

Communities throughout the United States are currently navigating hurricane season. Utilities are once again facing the enormous challenge of maintaining reliable electricity in the face of increasingly severe weather. Hurricanes bring more than strong winds and heavy rain. They can damage transmission lines, flood substations, isolate critical infrastructure, and leave millions without power for days or even weeks.

While utility crews continue to demonstrate extraordinary resilience and dedication during these events, a harrowing thought is on their minds: It is already time to plan for next hurricane season.

Siemens Energy opened the Grid AI Lab on April 13 in Orlando, Florida. The 6,000-square-foot facility combines NVIDIA AI infrastructure with Siemens Energy's deep grid expertise to develop and train advanced AI models that help utilities manage, operate, and strengthen their systems. The facility was built specifically to help utilities tackle real-world grid challenges through collaboration, experimentation, and innovation.

As this hurricane season moves through its peak months, Siemens Energy’s Grid AI Lab is the place to ask the critical questions that will help plan for the future.

• Which assets are most vulnerable?
• Which circuits should be inspected first?
• How quickly can service be restored?
• What infrastructure investments will have the greatest impact before the next major event?

Traditionally, answering those questions requires utilities to analyze data from numerous systems, conduct extensive field inspections, and rely heavily on historical assumptions. Today, artificial intelligence offers a new approach.

The Grid AI Lab is designed to bring together information from across the utility ecosystem. Data from grid-connected assets can be combined with aerial imagery, satellite observations, geospatial information, weather data, and advanced sensing technologies. By analyzing these diverse datasets together, utilities can gain a more comprehensive understanding of grid conditions and risks. The result is AI-driven insights that support planning, operations, resilience, preparedness, response, and recovery efforts.

One of the most promising applications is the use of digital twins.

A grid digital twin is a continuously updated digital replica of a physical asset or system — a substation, a transmission corridor, or an entire network. Unlike a static model, it continuously evolves as new data becomes available. At Siemens Energy, we are using NVIDIA technologies to create high-fidelity digital twins of power infrastructure. These digital environments enable utilities to visualize grid assets, analyze their condition, simulate future operating scenarios, and evaluate potential risks before they occur.

For hurricane preparedness, this capability can be transformative.

Imagine a utility being able to simulate the impact of a Category 4 hurricane on specific transmission corridors months before storm season begins. Engineers could identify vulnerable assets, evaluate alternative operating strategies, test restoration scenarios, and prioritize investments based on the highest-risk locations. Rather than reacting after damage occurs, utilities could proactively strengthen their systems based on data-driven predictions.

The same technology can help utilities during an active storm response. Digital twins can provide operational visibility, helping decision-makers understand changing conditions, assess infrastructure impacts, and allocate resources more effectively. Following a storm, AI models can be used to accelerate damage assessments and support restoration planning.

The value extends beyond hurricanes.

Many of the same technologies used to mitigate hurricane risks can also improve preparedness for floods, wildfires, ice storms, and other extreme weather events. While the specific data inputs may vary, the underlying principle remains the same: combining diverse data sources with advanced AI enables better decision-making before, during, and after major events.

What excites me most about the Grid AI Lab is that it is not simply a research facility. It was intentionally designed as a collaboration environment where utilities, technology providers, universities, researchers, and industry leaders can work together to accelerate innovation.

That collaboration is critical because the grid itself is changing rapidly. Electricity demand is growing. More renewable resources are being connected to the system. Customers expect greater reliability. Extreme weather events are becoming more disruptive. Meeting these challenges requires new tools, new technologies, and new approaches to planning and operations.

Resilience cannot be built overnight. Hurricane preparedness is no longer just about strengthening poles, wires, and substations. It is about strengthening decision-making. By combining AI, digital twins, and the intelligence framework of Noedra — Siemens Energy’s connected grid intelligence suite — utilities can gain a living understanding of their grid—anticipating risks, prioritizing investments, protecting critical infrastructure, and accelerating restoration before, during, and after the storm. The future of resilience is not built on isolated technologies, but on connected intelligence that helps utilities see farther, act sooner, and recover faster.

The investments utilities make today will determine how well they withstand tomorrow's storms.