An 80% Renewable Portfolio Standard in Alaska?

So, this one is a little different than my usual projects…

Back in December of last year, NREL got a request for a quick-turnaround analysis of a proposed 80% renewable portfolio standard (RPS) for the Alaska Railbelt electric grid, which stretches from Fairbanks through Anchorage to the Kenai Peninsula and provides roughly 75% of the state's electricity. The deadline was February 1st—giving us about 4 weeks to do the job.?

I wasn’t sure we could put together a model of that system in such a short time—but fortunately my colleague Marty Schwarz?had been working on a model of Homer Electric* (the southern part of the Railbelt system). (Marty was also a co-author of a great analysis?of extreme weather and the grid.) Plus, folks from NREL’s Alaska team (Elise DeGeorge, Sherry Stout, and Nathan Wiltse) were able to quickly connect with utilities, developers, planners, and agencies to get lots of data.?So, we were able to knock out a three-zone model in record time.?

We developed five 80% RPS scenarios, along with a no-new-renewables reference case, and did a bunch of the chronological simulations to check for load balancing and reserve violations while making sure we are hitting the RPS target.?We did a big-hydro case, a little-hydro case, and a few in between, with various amounts of wind, solar, geothermal, and tidal energy (a technology I’ve never modeled before…). We didn’t use a capacity expansion model, so we did everything iteratively, including resource adequacy estimates under big outage scenarios. I certainly wouldn’t want to do that on a system much larger than Alaska's (about 850 MW peak load in our 2040 scenario), but we got everything to work—and got some interesting results.

For me, the most interesting finding was the impact of the winter peaks, which normally scare me because I would typically rely on PV plus 4–6-hour duration storage to address summer peaks.?But not much sun in the winter meant relying more on wind and hydro, and the Alaska hydro system is more flexible than a lot of systems I am used to modeling.?Plus, with only an 80% RPS, we could keep plenty of the existing thermal fleet around (or add new thermal capacity) for peak conditions, or during outages of the Alaska Intertie between Fairbanks and Anchorage. We also knocked out the Kenai Intertie and a bunch of big plants simultaneously for extended periods during peak conditions to make sure we could keep the lights on. So, everything worked, and we were able to release the full report the same week the RPS bill was proposed.

We focused primarily on getting a better idea of what it might take to get the system to work technically, but we also considered some cost components, such as avoided fuel—and fuel in Alaska is quite expensive, despite how much of it they make up there! So I think that some of the scenarios will pencil out, but that’s going to require more analysis of the capital costs of some of these new hydro projects and other resources. So, if you want to know more about what it might take to get to 80% renewable electricity in Alaska, check out the report!

*Not to be confused with the awesome HOMER modeling tool developed by our former NREL colleague Peter Lilienthal.