Nov 16, 2023 · 58m · green-blueprint
AI in the real world: Solar forecasting, EVs, and virtual power plants
Carlos Nouel · 11m spoken Apoorv Bhargava · 10m spoken Jay Bombay · 8m spoken Stephen Lacey · 8m spoken Paul McDonald · 5m spoken Dr. Jack Kelly · 1m spoken Emma Woollacott · 1m spoken Erin Hardick · 1m spoken Dr. Noelia Otero-Felipe · 1m spoken News Commentator · 24s spoken Aidan McGivern · 16s spoken
0:00 / 0:00
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
This episode examines how machine learning and artificial intelligence are transforming the power sector, from dramatically improving solar and weather forecasting to orchestrating distributed electric vehicles and virtual power plants at the grid edge.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
The hosts as informed peer 4.3 Guest teaching 3.6 Guest disagreement 1.1 The hosts pushing back 0.1
speaking balance: gold is the hosts, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 55:40 Real-time rates turn consumers into Enron traders
Apoorv Bhargava emphatically rejects the push for real-time customer pricing, calling it a terrible idea that forces retail consumers to take on wholesale market volatility.
Hardest push from the hosts ▶ 32:09 Challenging the definition of EVs as DERsModerator Erin Hardick playfully calls out Apoorv Bhargava's provocative earlier remark that EVs shouldn't be treated simply as generic DERs.
Biggest teaching moment ▶ 6:55 Explain why solar forecasts create grid uncertaintyDr. Jack Kelly explains to Stephen Lacey that solar forecasting is the single largest uncertainty for ESO demand forecasts and details the architectural limits of numerical weather prediction.
The host holds their own ▶ 12:51 Diagnosing twenty years of power tech adoption lagStephen Lacey demonstrates deep industry knowledge by summarizing Jack Kelly's operational critique as the classic legacy barrier that has stalled grid technology adoption for over two decades.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The hosts as informed peer | Guest teaching | Guest disagreement | The hosts pushing back | Why |
|---|---|---|---|---|---|---|
| Extreme Heatwaves and UK Grid Vulnerabilities | 6 | 2 | 0 | 0 | Host Stephen Lacey narrates the narrative framing around UK extreme weather, interconnectors, and the symbolic return to coal, demonstrating strong domain context in his interview with tech journalist Emma Woollacott. | |
| Limitations of Supercomputer Weather Modeling and Episode Overview | 6 | 3 | 0 | 0 | Lacey and Dr. Jack Kelly discuss why numerical weather prediction falls short on solar forecasting, with Kelly explaining how machine learning on satellite imagery halves one-hour-ahead forecast errors. | |
| Analyzing Compound Extremes and Energy Droughts with AI | 5 | 4 | 0 | 0 | Dr. Noelia Otero-Felipe educates the audience and host on compound climate extremes and the emerging risk of summer energy droughts as air conditioning usage surges across Europe. | |
| Adoption Barriers and the Commercial AI Forecasting Race | 6 | 3 | 1 | 0 | Lacey synthesizes the commercial and institutional hurdles for grid AI adoption, noting that legacy IT requirements often trump pure model accuracy. | |
| Panel Introduction: Grid-Edge AI, EVs, and Virtual Power Plants | 4 | 3 | 2 | 0 | Apoorv Bhargava introduces WeaveGrid with spirited hot takes on how automotive data dwarfs utility data complexity and why EVs represent an unprecedented concentrated residential load. | |
| Customer Behavioral Engagement and Smart Meter Innovation | 3 | 4 | 0 | 0 | Carlos Nouel and Paul McDonald detail how utilities can drive consumer behavioral changes using simple proxy analytics before smart meters and high-resolution sampling are fully deployed. | |
| Optimizing Distributed Energy Loads and Wholesale Curtailment | 4 | 4 | 2 | 1 | Moderator Erin Hardick prompts the panelists on load management, prompting Apoorv Bhargava to argue against universal fixed-time EV charging and Jay Bombay to explain VPP wholesale curtailment optimization. | |
| Balancing Data Privacy, Consumer Trust, and Grid Value | 3 | 4 | 2 | 0 | Panelists explore consumer trust and privacy, with Paul McDonald urging strict cyber controls, Carlos Nouel stressing tangible consumer value, and Apoorv Bhargava playfully asserting the grid's emerging sex appeal. | |
| Mitigating Human, Machine, and Societal Biases in AI Models | 3 | 5 | 1 | 0 | The panel delves into AI bias risks, including legacy utility risk-aversion getting baked into predictive models, lost dispatch signals in VPPs, and socio-economic inequities favoring affluent EV owners. | |
| Five to Ten-Year Vision for Grid Edge and AI Integration | 3 | 4 | 3 | 0 | In closing visions, panelists outline edge-processing and automated program qualification, while Apoorv Bhargava forcefully condemns real-time pricing for retail consumers as turning individuals into Enron traders. |
Statements from this episode (22)
UK ran for 47 consecutive days without coal power before heatwave
“47 days the UK had been without needing any coal plants.”
Woollacott: Traditional weather simulations are constrained to one-kilometer resolution
“To do this properly, thoroughly it means an awful lot of computing power is required, and an awful lot of time, so the resolution can't be particularly high, we're looking at about a kilometre square, and they can't be particularly fast, because there's a huge…”
Kelly: Solar was National Grid ESO's biggest demand forecasting uncertainty
“The biggest source of uncertainty for their demand forecast came from their solar power forecast.”
Kelly: Numerical weather models struggle to generate detailed cloud maps
“Which is all to say that NWPs, whilst they do a great job of predicting things like humidity and temperature, converting that to a really detailed map of where the clouds are is known to be fairly imperfect at the moment.”
Open Climate Fix halved National Grid's one-hour-ahead solar forecast error
“We halved the error an hour ahead. At 24 hours ahead, I think the error was about 15% lower than their existing forecast.”
Climate change will shift grid energy droughts from winter to summer
“So we can expect, for example, more energy droughts in terms of high demand in summer than in winter, which is something that's before, you know, happened.”
Forecast accuracy ranks surprisingly low for transmission system operators
“The skill of the forecast is surprisingly low on the list of priorities when transmission system operators are buying forecasts. So it's more about how understandable is it? How reliable are you? Do you have a track record? Can you deliver the forecast in this…”
Lacey: ECMWF tested Huawei's 10,000x faster weather prediction AI
“Huawei has a three-dimensional weather prediction system that it says is 10,000 times faster than current methods, and it's been successfully tested by the European Center for medium range weather forecasts.”
Bhargava: EV home charging between 7-15 kW strains local residential distribution grids
“EVs are a completely new device type, each of them charging anywhere between seven to 15 kilowatts at home, and that is a tremendous amount of load to be putting onto a cul-de-sac that has eight homes, and now suddenly, 2.1 cars per household is starting to lo…”
Nouel: DER penetration peaks at roughly 20% even in leading markets
“Even in the places where we're really far ahead, like California and even Australia and other places, you know, penetration of DRs, whether it's EVs, solar, batteries, you know, heat pumps, you name it, it's still fairly low. I mean, it might make 20% at best …”
Nouel: National Grid tripled behavioral and EV program adoption using analytics
“And working through analytics, working through the data that we have, we actually have been able to actually triple, for example, adoption of the behavioral program and participation in the EV programs and participation in the weatherization programs.”
EV adoption will heavily shift grid cost-to-serve to local distribution
“There's two hundred eighty million cars in the country. There's a hundred and twenty million households. We designed a century old grid around those hundred twenty million households, and now as all these cars start going electric, there's just going to be a t…”
Bhargava: There is no universal optimal time for EV charging
“My controversial view is there is no right time for everyone to charge. There's only a right time for you to charge. And that you is not driven by just your location in the neighborhood and what time you decide to plug in. It's driven by how you choose to driv…”
AI models can identify specific real-time appliance usage from smart meters
“You know, the reality is because of the high level resolution that we're doing on the sampling and using language models, we actually stopped from, like, guessing what's happening in the home to, like, actually knowing what's happening in the home. Right to th…”
New York legally restricts utilities from targeted EV rate marketing
“Like, I can tell you, right now, here in New York, utilities couldn't go and just market. If I know you have an EV, I couldn't just go and market people saying, here's a great rate for you that can save you money. Even though it's a good thing for customers, t…”
Utility sector faces first electricity demand growth since air conditioning invention
“The truth is, suddenly, the utility industry is going from 40 years of flat and declining load to the first growth moment since the air conditioning unit was invented.”
Bombay: Utility and regulator trust is the biggest roadblock for DER proliferation
“It's probably one of the biggest challenges to the DER industry, which is trust from the utility and regulator side on embracing DERs and really accepting them as a reliable resource, whether it's for capacity or for localized capacity I think getting, establi…”
AI energy models risk perpetuating human over-planning and forecasting biases
“And I think the challenge with those models and those biases getting into the models is that all the things that we're trying to get away, those inefficiencies we all talk about getting away from, like over-planning, over-forecasting, having over-capacity on c…”
Distributed energy rate designs risk favoring the wealthiest ten percent
“I mean, a lot of people who buy these distributed resources today EVs very much being that, solar, whatever, are generally very wealthy people, and so are we now doing rate design and setting operational constraints based on what the richest 10% want, or are w…”
McDonald: AI can auto-enroll low-income consumers in clean energy assistance programs
“I think it's entirely possible to use artificial intelligence, to use, ah, layered data sets to predict. In the same way that we get notifications that we're pre-qualified for a loan or a higher credit card rate, we're pre-qualified for income-eligible financi…”
Bombay: VPPs will resolve localized grid bottlenecks within 5 to 10 years
“I think VPPs have quite a unique role to play in helping alleviating those very localized pain points. And we're not even there right now. We're just talking system wide capacity is pretty much like the limit as to how much VPPs have been able to contribute in…”
Real-time electricity pricing is a terrible idea for retail consumers
“I also hear a lot of things about, we're magically gonna give everyone real-time rates. I think that's a terrible idea. I think that is a terrible idea. I think real-time rates are a, are an approach to trying to personalize, but actually passing the entirety …”