Hear Portland General Electric’s Larry Bekkedahl in a fireside chat, as we discuss the role that data is playing in helping the utility achieve “Net Zero by 2040” in the on-demand session from UA Summit 2021 called “Unlocking Operational Outcomes: Digital Transformation and Clean Energy Initiatives with AMI Data.”
It won’t surprise you that as part of its “Net Zero by 2040” goal, Portland General Electric (PGE) is investing in clean and renewable energy generation, including wind, solar and hydro power. But the utility is also making another significant investment that they deem just as vital to their net zero success: Data.
“Utilities are thinking about decarbonization and how we’re going to meet decarbonization targets globally. We’re all going to be asked to move very fast in this space, but it takes data, it takes information,” explains Larry Bekkedahl, PGE’s Senior VP, Advanced Energy Delivery (previously VP Grid Architecture, Integration & System Operations). “Now that we have smart meter data, we need to take that data and turn it into knowledge and decision making tools.”
PGE believes that data drives four critical elements in the net zero equation:
- Helping customers see their contribution and impact on PGE’s net zero goal
- Enabling targeted and personalized marketing and offers for each customer rather than blanket efforts
- Making possible virtual power plant and load balancing
- Facilitating flexible feeders – influencing peaks, dynamic pricing, shifting load management
Empowering Customers to be Part of the Clean Energy Revolution
“Helping our customers think about how they achieve decarbonization is really important,” emphasizes Bekkedahl. “We need to empower our customers. We need to provide them with the information and knowledge they need so that they can react and make the right decisions.”
PGE is sharing disaggregated energy use data with its customers’, explaining their energy use signatures and how they contribute to carbon reduction. With that understanding, the utility can then ask its customers to play an active role in their collective clean energy journey by taking actions such as:
- Buying electric vehicles and charging them at off peak times
- Installing rooftop or adopting community solar, reducing grid reliance
- Shifting and reducing energy consumption to minimize peak demand through energy efficiency and automated demand response
- Replacing gas and inefficient appliances with electric efficient appliances, to advance electrification at scale
As utilities transition to dynamic customer engagement strategies like this one, Bidgely’s UtilityAI™ and its AI-powered AMI energy disaggregation enables utilities to deliver multi-channel, personalized alerts, recommendations and other communications that generate greater customer energy and cost savings, inspire behavior change and position utilities as trusted advisors.
Personalizing Customer Outreach
Before the availability of smart meter data, utilities relied upon substation data or customer participation in home audits. They might have also leveraged local government records to see what appliances might be installed in a home. But the data was limited and it wasn’t real time or seasonally specific. As a result, most utility marketing was blanketed to entire service territories.
The problem with blanket marketing was two-fold. First, without insight into a household’s unique energy habits, preferences and usage patterns, utilities communications were frequently not relevant or timely to individual customers. Secondly, when it comes to making better energy decisions, there is no universal motivator or set of rules that applies to all customers, making it impossible for any blanket call to action to have a meaningful impact
UtilityAI’s AI-powered customer segmentation transforms blanket marketing to micro-targeted, personalized engagement allowing utilities to replace generalities with precise and profoundly relevant energy saving insights and advice that more effectively influence behavior.
“We now know which customers can reduce their cost by putting in a different heating system or hot water heater or better controls,” says Bekkedahl. “Now we can target individual customers instead of just blanketing messages out to all of our customers.”
The availability of actual home energy use data is empowering the marketing team to be more efficient, while also fostering greater collaboration with the grid group.
“If you think about utilities, usually the energy efficiency group was over on this side. And the operations folks were over on this side. And there was a wall in between the two and they didn’t really talk to one another,” agrees Bekkedahl. “With what we’re doing today with data, we’re bringing those groups together. From a total company perspective, we have both sides of the equation involved and working together to actively pursue what we can do to change.”
Achieving Personalized Flexible Demand
“In the past, generation followed demand, and demand response played only a small role in curbing peak demand. But with renewables, we have to figure out how to make demand follow generation,” says Bekkedahl. “Now that disaggregation tells us what is happening in the house, we have to influence consumers to achieve flexible demand. Data can help. Being able to influence when the heating and cooling is on, or when an electric vehicle is charging is going to be really important as we go to the future.”
Historically, it has been common for utilities to use time of use rates to shift customer energy use outside of peak times. Typically, that meant charging customers most during the middle part of the day when temperatures were highest. As such, utilities would incent people to conduct their electrically intensive activities in the mornings or in the evenings, and avoid the middle of the day. However, the addition of more solar generation meant changing use patterns to encourage activity in the middle of the day when solar generation is highest, and discourage activity in the morning and evening.
“TOU really makes a difference because it does motivate customers to think about when they should use energy. But TOU rates can also create problems,” cautions Bekkedahl. “If we have a neighborhood in which six homes have an electric vehicle, and if they are all on a TOU rate that triggers at 10 p.m., you’re probably going to overload the transformer that serves those homes when they all try to charge at the same time. So we have to think differently about demand now as we go into the future.”
Bidgely’s UtilityAI facilitates TOU granularity and allows utilities to see each customer’s home charging behavior in real time and encourage charging at off-peak times, even as off-peak times shift with evolving generation.
“We’re trying to match loads where there is capacity in the system. So if we have excess capacity during certain parts of the day, we want to incent people to charge during those periods. If I see a constraint on a feeder, I hopefully can influence enough of that load,” Bekkedahl explains. “We as operators are going to think differently about how we operate at a feeder/substation/distribution grid level, and household energy use data is critical to understanding what the flexibility is. We collect data from the meter, and then roll it up to the feeder, substation and total-balancing level. As the operator of a grid, I am going to have to build a lot more capacity into the system if I don’t have that flexibility, and it’s not going to be nearly as affordable.”
Alleviating Load with Distribution Management
“Knowing when customers are using energy and what customers are consuming opens up the potential to take service endpoints and virtually assign them to a different feeder to alleviate the constrained feeder,” says Bekkedahl.
With data from individual home and small business meters, UtiltyAI Analytics Workbench allows utilities to break usage down to the appliance level, and then aggregate that data back to the feeder level or zip code level or any kind of geographical distribution level.
Analytics Workbench’s AMI-data-based tools allow utility teams to “slice and dice” energy data to improve grid planning by pinpointing where load is increasing — such as by revealing how many people are buying EVs and in what areas. The result is more reliable, data-informed decisions about where to expand capacity and where, when and how to shift the load from one feeder to another.
“When I started in this industry, we’d analyze the load on a five-year basis to see where the peaks were, calculate growth and make estimates to guide the build out of the distribution system to match the peak. Because things weren’t moving that fast, five years was adequate unless you knew that there was a big load that was coming in,” recalls Bekkedahl. “From five years, we went to annual, then to seasonal, then to daily, and then to hourly. Now we’re down to real time — basically every three seconds we’re doing a test of the system. And that tells us what’s happening moment to moment, but it also helps us to understand what is likely to happen an hour from now, three hours from now, or tomorrow. And I see the same thing in distribution.
“I like to refer to this way of doing business as the virtual power plant. We’re moving now to a digital environment and forecasting for tomorrow. And for example, knowing what the temperature is going to be tomorrow, I can now plan what to turn on or off to influence the system. And if I can see that I have a problem coming, I know that I am going to have to take certain actions. Today, when we think of transmission and generation, we think of thousands of points. And when we’re thinking about distribution, we’re thinking about millions of points. Analytics Workbench would really help us, because it gives us robust data about flexible load that allows us to make those essential real-time decisions.”
Hear more from Larry Bekkedahl about the role that data is playing in helping the utility achieve “Net Zero by 2040” in the on-demand session from UA Summit 2021 called “Unlocking Operational Outcomes: Digital Transformation and Clean Energy Initiatives with AMI Data.”