Visit your regional NVIDIA website for local content, pricing, and where to buy partners specific to your country.
Enhancing power generation, transmission, and distribution for grid modernization and lower costs.
Building a software-defined smart grid enhances resiliency during extreme weather events, improves electricity affordability and reliability through real-time optimal power flow, integrates distributed energy resources such as solar, wind, electric vehicles (EVs) and home batteries, and accelerates decarbonization initiatives. With AI, utilities can modernize legacy grid infrastructure, scale renewable energy, and deliver lower cost power to homes and businesses.
Learn how AI and edge computing can help utilities forecast energy demand, dynamically manage supply, identify real-time outage risks, and improve customer experiences in contact centers.
Learn how Utilidata, Hubbell, and NVIDIA are bringing distributed AI to the grid edge, giving utilities real-time visibility into grid operations and distributed energy resources.
Managing Distributed Energy Resources
Managing distributed energy resources, such as solar panels, wind farms, electric vehicles (EVs), and home batteries presents challenges for traditional grids. Edge AI helps dynamically manage these resources, predict demand, and allocate supply to enhance grid resiliency. Advances in smart meters—powered by a software-defined smart grid chip based on the NVIDIA® Jetson™ edge AI platform—deliver greater value to utilities and their customers, while unlocking new opportunities for clean energy companies and third-party market developers.
Automating Grid Asset Inspection
Most drone inspections still require a human to manually inspect the video for defects. Training a computer vision model to automate inspection is difficult without a large pool of labeled data for every possible defect. By using synthetic data generation in Omniverse, utilities can automatically create thousands of labeled, photorealistic examples of diverse defects. These synthetic images can then train an inspection model for real-time drone inspection.
Optimizing Power Flow and Outage Scheduling
The emergence of microgrids and nanogrids—small-scale local backup power systems that operate autonomously and integrate distributed energy resources—has made real-time load balancing and price adjustments exponentially more complex. High-performance computing (HPC) lets utilities quickly model networks at scale and achieve optimal power flow for outage scheduling and contingency analysis. AI-enhanced intelligent video analytics and cybersecurity secure grid infrastructure from unauthorized access.
Analyzing Infrastructure to Automate Repairs
In the U.S. alone, utilities own 185 million poles and spend tens of millions of dollars each year tracking the status of crossarms, transformers, fuses, and other grid assets for faults. This manually intensive process can take up to a decade, yet the condition of each device is critical to delivering power safely to homes and businesses. By deploying the NVIDIA Jetson edge AI platform on field service trucks, utilities can automate data collection and analysis, accurately monitor real-time grid health, and identify needed repairs and vegetation management.
Personalizing Customer Interactions
Customer service, a key metric for utilities, relies on 24/7 service availability. Higher-than-normal demand volatility due to unexpected outages can cause surges in call volume from affected customers. However, AI and data analytics enhance the overall customer experience and help control operating expenses. In call centers, utilities can use AI as virtual assistants to support agents. Data analytics can be utilized to present the customer support agent with a 360-degree view of the customer, including past usage, peer comparison, and rate guidance, which reduces call times and customer frustration.
Improving Truck Rolls During Extreme Weather
During power outages due to heavy wind, rain, or other weather events, utilities are expected to quickly and efficiently deploy fleets of field service trucks to restore power to residents and businesses. Utilities are exploring using AI to optimize vehicle route planning in real time, reduce travel times, cut fuel costs, and improve response times.
Learn how AI is improving power generation, reducing energy costs, and advancing scientific breakthroughs.
Exelon, the largest regulated electric utility in the U.S., is using NVIDIA Omniverse to automatically create thousands of grid asset defect examples for real-time drone inspection models—enhancing grid reliability and resiliency.
Minerva CQ, a member of the NVIDIA Inception program, is making customer service calls in energy quicker and more efficient for agents and customers. Learn how the startup uses NVIDIA Riva for real-time dialogue suggestions, sentiment analysis, and optimal journey flows.
Image courtesy of Open Climate Fix.
Open Climate Fix, a member of NVIDIA Inception, built transformer-based AI models trained on terabytes of satellite data to improve solar energy generation predictions by 3x. Learn how the nonprofit is helping decarbonize the UK's electric grid.
We had collaborated with NVIDIA going back to our purchase of an NVIDIA DGX-1 system a few years back... We used NVIDIA Omniverse to generate different defects of cross-arms, which would produce labeled data for the training of the inspection model.
— Vladyslav Anderson, Principle Quantitative Engineer, Exelon
By using synthetic data generation in NVIDIA Omniverse, our goal was to automatically create thousands of labeled photorealistic examples of various defects in grid assets. We are in the process of using real images and these synthetic images to train inspection models.
— Justo Ankush Agarwal, Director of Advanced Analytics, Exelon
In Oregon, we are experiencing the impacts of climate change first-hand and recognize the urgent need for innovation at the grid edge as we transition to a clean energy future. Investing in new technologies for the grid is a key strategy for PGE to achieve its climate goals and provide customers with clean, affordable, and resilient energy.
— Ananth Sundaram, Senior Manager of Integrated Grid, Portland General Electric (PGE)
With the increase in distributed energy resources–especially batteries, solar, and electric vehicles–those technologies can pose a management problem or opportunity to utilities. We’re working with NVIDIA to move that central computation out to the edge of the grid, so we can manage those resources in real time and integrate them into real-time power flow dynamics from the substation to the end of the line.
— Marissa Hummon, Chief Technology Officer, Utilidata
By combining Deloitte's deep data science and energy industry experience, NVIDIA's accelerated computing and AI platform, and Utilidata's Karman platform, the new initiative provides intelligent grid solutions for utilities to utilize renewable energy sources with real-time, actionable insights.
Schneider Electric announced a collaboration with NVIDIA on new reference designs to optimize data center infrastructure for performance, scalability, and energy efficiency—paving the way for groundbreaking advancements in AI and digital twins.
Image courtesy of Crystal Group.
Aclara will be the first company to embed Utilidata’s distributed AI platform, Karman, in a smart meter to enable a connected grid that delivers clean and reliable energy. Built on a custom NVIDIA module that leverages AI, Karman captures robust, high-quality data to improve grid operations and manage distributed energy resources.
Image courtesy of Utilidata.
Learn about the AI and HPC hardware, software, and networking solutions for utilities.
NVIDIA Jetson brings accelerated AI performance to the edge in a power-efficient and compact form factor. Together with the NVIDIA JetPack™ SDK and NVIDIA Isaac™ software for Robotics Operating System, these Jetson modules, including the new NVIDIA Jetson Orin Nano™, support a full range of edge AI and robotics applications.
With NVIDIA AI Enterprise, energy companies can speed up development of use case applications, such as reservoir simulation, seismic processing, demand forecasting, predictive maintenance, and power grid management. Learn how to get access to NVIDIA AI with free curated labs to start testing.
NVIDIA DGX Cloud is a multi-node AI-training-as-a-service solution optimized for the unique demands of enterprise AI. Access NVIDIA DGX Cloud to experience a combined software and infrastructure solution for AI training that includes a full-stack AI developer suite, leadership-class infrastructure, and concierge support.
NVIDIA Fleet Command™ is a managed platform for container orchestration that streamlines the provisioning and deployment of systems and AI applications at the edge. It simplifies the management of distributed computing environments with the scale and resiliency of the cloud.
NVIDIA cuOpt™ is an optimization API that uses AI to help developers create complex, real-time fleet routing. These APIs can be used to solve complex routing problems with multiple constraints and deliver new capabilities—like dynamic rerouting, job scheduling, and robotic simulations—with subsecond solver response time.
NVIDIA Riva is a GPU-accelerated speech AI SDK for building and deploying fully customizable, real-time AI pipelines that deliver world-class accuracy in all clouds, on premises, at the edge, and on embedded devices.
NVIDIA Morpheus is a framework that lets cybersecurity developers create optimized AI pipelines for filtering, processing, and classifying large volumes of real-time data. Bringing a new level of security to the data center, cloud, and edge, Morpheus uses AI to identify and act on threats and anomalies previously impossible to identify.
NVIDIA Metropolis features GPU-accelerated SDKs and developer tools that help developers optimally build, deploy, and scale AI-enabled video analytics and IoT applications—from the edge to the cloud.
NVIDIA Omniverse is an extensible, open platform built for 3D virtual collaboration and real-time physically accurate simulation. Omniverse combined with NVIDIA Modulus, a framework for developing physics-informed machine learning (physics-ML) models, enables digital twins for wind farms, power plants, electric grids, and, someday, Earth itself.
Videos
Webinars
Use Jupyter iPython Notebooks on a Jetson Nano Developer Kit to build a deep learning classification project with computer vision models. This easy-to-use, powerful computer runs multiple neural networks in parallel.
Gain an understanding of the various building blocks of NVIDIA Modulus, the basics of physics-informed deep learning, and how the framework integrates with the overall Omniverse platform.
Find out how to model mixed-fleet vehicle-routing problems for delivery or intralogistics and apply multiple routing constraints like capacity, time windows, mixed priority, or skills with simple APIs.
Explore NVIDIA Inception, the free program designed to help startups evolve faster through access to cutting-edge technology and NVIDIA experts, connections with venture capitalists, and co-marketing support to raise visibility.
Connect with millions of like-minded developers in the NVIDIA Developer Program to do your life’s work, and gain access to free containers, pretrained models, SDKs, technical documentation, and peer and domain expert help.
Expand and support your portfolio with NVIDIA’s Venture Capital (VC) Alliance, an initiative between NVIDIA and investors around the world who are focused on startups building cutting-edge technology with AI, data science, and HPC.
NVIDIA solutions for the power and utilities industry go beyond products. Our partners are here to assist your organization at every level to build and execute transformative AI strategies, products, and services.
Chat with NVIDIA energy experts to help solve your business challenges.
NVIDIA Privacy Policy