Demand-Side Flexibility and Electricity Use: Q+A With Dr. David Brown

Dr. David Brown, Energy Economist at the University of Alberta

In February 2024, Dr. David Brown, energy economist at the University of Alberta, spoke with Future Energy Systems’ Research Communications Coordinator Elyse Dzenick about electricity market pricing structures. The conversation shed light on David’s ongoing research at the University of Alberta, in partnership with scholars from the University of Calgary and Stanford University, as well as industry partners. The work focuses on supporting Alberta’s electricity grid as the availability of renewable energy resources increases, along with growing electricity demand. 

Q: Who’s involved in this research?

A: My co-authors are Drs. Megan Bailey, Blake Shaffer, and Erica Meyers at the University of Calgary and Frank Wolak at Stanford University. We have four utility players, and more are involved in different ways across Canada. 

Q: Can you briefly define demand-side flexibility?

A: Electricity supply and demand must be perfectly matched at all times. Historically, we’ve relied primarily on just turning on power plants like natural gas facilities to meet this demand. But with the increasing amount of renewable generation, which results in more variability on the system, we’re looking for more resources rather than just calling on natural gas facilities, can we leverage something else that exists in the market?

So there’s an increasing emphasis on demand-side flexibility, which is something as simple as saying “Can I, as the supplier, call upon the customer to reduce their usage at home?” So think about delaying the washer and dryer, turning off lights, and avoiding charging their electric vehicle. That right there is the demand side flexibility and a lot of folks ask how can we unlock the demand side of the market rather than just relying on the supply side.

Q: Can you share examples of demand-side flexibility in action?

A: Yeah. Think about the emergency alert that I think everyone in Alberta got on their cell phones [in January 2024] when we had that extreme cold snap. Effectively, the government was calling upon all Albertans to lower our electricity demand to reduce the strain on the system. But I think there’s a much broader scope of demand-side flexibility, which is basically a structure saying: “I’m going to charge you more money to consume electricity from 5:00 pm to 10:00 pm.” If you can shift your demand outside that window, I’m going to create incentives for you to do that. That is a really common pricing structure to motivate demand-side flexibility. 

Q: Can you go into what critical peak pricing (CPP) and time-of-use (TOU) actually means and how they relate to your research?

A: Those are two common pricing structures in North America. Time-of-use pricing is very predictable. Basically, every single day from 5:00 pm to 10:00 pm, I’m going to charge you more money for your electricity consumption, and then from 10:00 pm to 6:00 am, I’m going to charge you less money. The idea is that from 5:00 pm to 10:00 pm, that’s when the grid is under the most pressure. So it’s very predictable –– just happens every day and people can respond to it. The classic example in Canada is Ontario’s market, and it definitely has its benefits. 

On the other hand, CPP relies on these infrequent events. Say that on Wednesday, it’s going to be cold in Edmonton and there won’t be much solar or wind generation. I’m going to send out an alert, through the utility app or by text message, that calls upon you to lower your demand. And in that one-shot window, I’m also going to charge you a higher price. It’s Alberta’s current electricity market pricing structure. 

We’re focusing on TOU pricing and our focus group involved participants using an app that we helped design to lower their electricity demand. We’re really interested in leveraging automation that can motivate you to lower electricity use or even do all the work for you automatically. At the end, we’re trying to understand how to best shift demand in those two ways. 

Q: Your research compares these pricing structures with automated systems to see what works best for demand-side flexibility practices. Have you seen that people respond better to one model over another?

A: I would say that the key is to make things as simple as possible. With CPP alerts, people will respond the first couple of times, but then over time, people get busy and become less attentive. Our work is trying to evaluate the relative success of pricing signals versus just automating the entire system. 

So automating your thermostat, your electric vehicle charger and your hot water heater make them automatically respond to these [critical peak pricing] events and you don’t have to do anything. We’re finding that the automated systems are resulting in much larger demand reductions than actually sending out alerts to unplug or delay use. Basically the “set it and forget it” approach is what we’re finding very successful.

Q: What challenges have you come across with demand-side flexibility and where could the practice cause problems down the line?

A: I think one big challenge, and this is where people critique demand-side flexibility, is that there’s some variability in how customers are going to respond. They’re not going to respond perfectly every day. And you can’t really dispatch their responses the same way suppliers can dispatch electricity. What we’re finding in our research is that with technology and automation, you get a very predictable and consistent level of demand response, rather than price-based signals, which overcomes some of the downsides of demand-side flexibility.

In some cases, like with TOU in particular, the cure can actually be worse than the disease. By getting people, for example, with electric vehicles (EVs) to shift charging from right when they get home from work to 10:00 pm –– to when the price decreases –– we’re still going to get a really big spike in electricity demand on the local neighbourhoods when they’re plugging in their EVs. So all we’ve effectively done is shift the peak demand later at night. But we’re using this information to improve. Instead, we’ve learned to set the automation to different intervals. So there’s a way to enhance DSF by leveraging that information and technology in a smart way. 

A second challenge is education, particularly for EV owners. We have to make sure EV owners have a charged car when they need it. For this, the owners set two parameters: the target charge percentage (how much you want to charge your battery), and when you want to leave. Those two factors serve as hard algorithmic constraints. So owners can keep their cars plugged in the whole night, and we can manage the charging and guarantee the needs of EV owners are met. And we’ll pay you money for helping the grid, and you still have a lot of flexibility in what you use. 

Q: What kind of results have you seen?

A: There’s been a lot of really good empirical work looking at implemented TOU pricing structure or a CPP structure and the amount that people will respond to that price signal. And typically, unfortunately, they find what we call in economics “demand elasticity,” which is basically saying for a 10% increase in price, typically you only get 1% to 2% reduction in demand. So what we’re finding is that people respond to price, but not very strongly. 

What we find is that emerging technologies really help with getting a larger demand response. During CPP events, historically it’s been about a 5-10% reduction in demand. We’re finding as high as 25-35% reduction during these events, just by installing automated systems that control energy use. And consumers do have the option to opt-out at the time of the event if they want.

It works like this: we’ve developed an app that can either set a schedule or you can give the utility companies access to your usage via the app and they can manage power use based on demand. And there’s a button that can stop that at any point if you feel uncomfortable. And for doing this, we’ll pay you money, maybe $10 to $20 per month for allowing us access to this. 

Q: Once your research is complete, what happens next with the results? 

A: I’d love for some of these methods to be possibilities in the short term, but the reality is that some regulatory policies and hearings take years to pass. So right now we’re focused on building evidence for the success of different DSF programs that we can show to policy regulators and system operators for decreasing electricity demand. 

And we’ll share the results more broadly. First are the academic publications, of course, because that’s the requirement. We’re also working with utility companies, who are using our findings to inform their formal reports to the Alberta Energy Regulator and other decision-makers. 

____

David’s interview is also featured on our Youtube: 

Demand Side Flexibility: Dr David Brown 

Expanding on Demand Side Flexibility: Dr. David Brown 

This blog is part of an ongoing series promoting Future Energy Systems’ Accelerator Projects.