Energy Future: Powering Tomorrow’s Cleaner World
Energy Future: Powering Tomorrow's Cleaner World" invites listeners on a journey through the dynamic realm of energy transformation and sustainability. Delve into the latest innovations, trends, and challenges reshaping the global energy landscape as we strive for a cleaner, more sustainable tomorrow. From renewable energy sources like solar and wind to cutting-edge technologies such as energy storage and smart grids, this podcast explores the diverse pathways toward a greener future. Join industry experts, thought leaders, and advocates as they share insights, perspectives, and strategies driving the transition to a more sustainable energy paradigm. Whether discussing policy initiatives, technological advancements, or community-driven initiatives, this podcast illuminates the opportunities and complexities of powering a cleaner, brighter world for future generations. Tune in to discover how we can collectively shape the energy future and pave the way for a cleaner, more sustainable world.
Energy Future: Powering Tomorrow’s Cleaner World
How AI is Revolutionizing the Grid: Efficiency, Reliability, and Resilience
Uncover the significant role of generative AI in revolutionizing power grid modeling and price forecasting. Discover groundbreaking work by the National Renewable Energy Laboratory, where advanced neural networks are accelerating the design of wind turbine blades and simulating complex grid scenarios. These innovations could predict the impact of electric vehicle adoption on energy loads and infrastructure, marking a step towards a more sustainable future. As we embrace these technological advances, we also address the need for balance, considering the additional demands AI places on the grid. Stay tuned for our next session, where we'll focus on AI's transformative effects on the distribution grid.
In the past two weeks, since I put up the last of five videos on AI data center-driven demand on the grid, there have been a number of critical updates, with perhaps the most interesting being that Meta recently announced they will issue an RFP seeking 1 to 4,000 megawatts of new nuclear generation to come online in the 2030s. Winning that bid would certainly boost the investability of any number of startups in the space, although shares of the publicly traded modular nuclear reactor companies haven't been doing too badly lately. But let's turn our attention away from the data center's near-insatiable appetite for juice and discuss ways that AI could actually help to make the grid more efficient, reliable and resilient. A good rule of thumb about AI in this space is to think about the following questions what could I do in various areas across the grid if I had access to instantaneous, accurate and highly detailed information? What if I had the ability to ask any question about the grid and grid-connected assets that I wanted? And what if I could reliably and instantly act upon that information and receive real-time feedback in response to my actions?
Speaker 1:To better understand AI's potential value proposition here, let's start with the basic challenges and opportunities to be addressed. First, our grid is bursting at the seams both at the transmission and distribution levels, with infrastructure increasingly hitting thermal limits at certain times, largely when ambient temps are high and there's a lot of current flowing. Second, there's an enormous pent-up demand for new supply, with difficulties in getting infrastructure permitted and built. And third, the grid is increasingly vulnerable to severe weather. And fourth, it's still carbon intensive.
Speaker 1:Now let's look at how AI-assisted tools can help. Today, we'll tackle some promising use cases on the supply side in the bulk power system, with transmission, ai holds significant potential. For example, it can help with predictive maintenance, with algorithms scanning data sets to look for anomalies that indicate potential looming equipment failures. Operationally, ai can help boost the performance of transmission lines by assisting certain grid-enhancing technologies known as GETs. These don't involve building out new infrastructure, but rather making more efficient use of the equipment we already have. The first of these is dynamic line rating, in which historical method of limiting lines based on static capacity ratings is ditched in favor of an approach looking at actual ambient conditions. Lower temps and higher wind speeds, for example, pull heat from transmission lines, allowing them to move more power, in some cases as much as 50% or more. That helps limit congestion bottlenecks and aids with the interconnection of more generating assets, rather than being subject to a firm 24 by 7, 365-day line rating that would preclude their interconnection. They might be able to operate most of the time and only be curtailed a few hours during the hottest and stillest days. Absent the situational awareness afforded by AI, you could not adopt this flexible and pragmatic approach. This may be most helpful to wind assets since, logically, during the same periods when wind turbine output is high, that same wind is dissipating heat from transmission lines. Then there's so-called topology optimization, a fancy phrase for directing electron traffic by opening and closing breakers to route power differently, facilitating higher utilization of assets. Ai can help here as well by more quickly assessing a wider variety of scenarios across the grid, and this can also help with interconnecting new assets, speaking of which interconnection itself is a big problem.
Speaker 1:In its latest report, lawrence Berkeley Labs noted that the median time for interconnecting new assets from the study request to transmission operators to the time power actually flows is north of five years. In the year 2000, it only took two years to interconnect because planners were dealing with fewer and far larger projects, mostly big coal and gas plants, the elephants. In large part, today's delay is a result of a changing resource mix and a numbers game. The elephants have been replaced by cats and dogs, and rats and frogs, with large numbers of projects as small as 10 megawatts joining the queue. In the 2000 to 2004 period, about 300 projects waited patiently in the queue. That number grew to about 850 annually from 2005 to 2014, and has since soared to over 3,000 requests per year in recent years, with now 10,000 projects impatiently languishing in that queue. The challenge posed by these large numbers is that there are many more moving parts to evaluate. Each new project or potentially approved cluster of projects changes the situation for those projects waiting downstream, creating a constantly shifting dynamic that dramatically increases the need for planning resources and sheer computational capability. This is an area where AI holds enormous amount of potential in both cutting the time required and increasing the number of scenarios that can be assessed in that interconnection planning process. The DOE reports, the transmission owners and software developers have started to deploy these newer models, but substantial work remains to be done in this area.
Speaker 1:At the same time, a lot's happening on the generation side of the equation, with numerous opportunities and digitalization of various assets that will help operation of the gen fleet and dispatch strategies. Gas generators, for example, can be run more efficiently based on actual operating conditions rather than prescribed schedules. For example, algorithms applied to information derived from a variety of sensors can tell the plant operators how hard they can run a turbine, perhaps overfiring and exceeding nameplate ratings on a frigid day when demand is high and power prices are soaring. They can also better understand when to take turbines out for maintenance rather than relying on fixed schedules. Perhaps an apt analogy here would be changing the oil in your car, not based on the miles driven or months since your last visit to Jiffy Lube, but based on the sensors in the oil actually telling you how dirty that oil is.
Speaker 1:Ai, married to more powerful computers, is helping to generate more accurate weather forecasts. Longer term and more geographically precise locational forecasts can help great operators refine their output projections and dispatch strategies, while optimizing utility-scale battery storage and dispatch as well. Within a wind farm, ais can also help with so-called wake steering, which has nothing to do with directing mourners towards an open casket. No, by orienting wind blades a few degrees, computers can minimize the disruptions in wind flow affecting downwind turbines, thus optimizing output. Among other benefits. The National Renewable Energy Laboratory, nrel, suggests that integration of wake steering strategies into the siting and planning process could cut land requirements for future wind plants by an average of 18% and up to 60% in some instances. Ai can also help new sustainable technologies such as advanced geothermal projects that extract heat from solid rock miles underground and use that to generate power. Here machines and algorithms do all sorts of things, from telling operators where to drill to physically guiding the drill bits through the subsurface hard rock, predicting reservoir behavior and determining how much heat to extract from a given area over a specific duration. In addition, ai has the ability to speed up environmental reviews and completion of documentation for interconnection requests the boring but necessary stuff associated with these projects.
Speaker 1:Some of these applications are already happening with AI related to machine learning.
Speaker 1:Here, computers take existing information and make sense of it using algorithms that identify patterns to make decisions.
Speaker 1:But as the large language models become increasingly powerful and more sophisticated, the ability to develop generative AI to understand the patterns of existing data and then generate new data to improve decision making well, that will take us to the next level. Here we'll see increased abilities to model the nation's power grid, predict future power prices, develop better technology. Nrel reports, for example, it's already employing advanced neural networks to improve the design of wind turbine blades 100 times faster than through previous methods, and model countless what-if scenarios across the grid. For example, what if EV adoption increased by X percentage in a specific location? What would the impacts be on hourly loads and associated supply infrastructure? That's where we're headed. So not only should we get used to AI, we should lean into it. If these data centers and their large language learning models are going to stress the grid on the demand side, we might as well get as much value out of these new capabilities as we possibly can. In our next session, we'll talk about the impacts of AI on the distribution grid. Thanks for watching and we'll see you again.