Heat pumps a possible solution for rural wind farms
By Maisie Thomas
On one of Nome’s many windy days, Banner Wind Farm can generate up to 35 percent of the town's power. The yearly average, though, is much lower, at nine percent. This is in part because Nome Joint Utility System is unable to take full advantage of the amount of wind the region receives.
Researchers from the University of Alaska Fairbanks are studying how rural utilities like Nome’s could make more use of their wind by pouring excesses of it into household heat. They could begin demonstrating the system in May and hope to start testing the model with customers afterwards, according to Andrew McDonnell of UAF’s College of Fisheries and Ocean Sciences, one of the two researchers studying the problem.
On the one-to-seven scale used by the wind power industry, most of the Bering Strait region gets Class 4 to Class 7 winds, according to the Alaska Energy Authority. For utilities that must balance variable wind energy with other generation — usually diesel — the problem is often not too little wind but too much.
In a 2015 blog post, Alaska Center for Energy and Power Director Gwen Holdmann describes an irony facing bush Alaskan wind farms such as Banner. Since there is no electrical transmission grid connecting rural communities, the energy the turbines produce must be consumed immediately, even if the demand is lacking at the time. “The energy must be consumed where it is produced, and when it is produced,” as Holdmann puts it.
This creates a challenge for utility managers such as NJUS’s John Handeland. Regulating a system powered by a mixture of diesel and wind requires a balance of technical and regulatory factors. An important example Holdmann provides is that NJUS’s largest diesel units cannot be run below 50 percent capacity. Combustion engines are less efficient at lower outputs, so running the generator at a lower capacity would put NJUS out of compliance with Environmental Protection Agency standards. Because the wind is not always blowing when energy demand is high and, conversely, when demand is high there may be no wind, the result is an “ironic situation where John [Handeland] needs to spill wind power and burn diesel when his loads are low and wind speeds are high,” Holdmann explains.
A solution may be in the works, however. University of Alaska Fairbanks faculty members McDonnell and Jeremy VanderMeer with the Alaska Center for Energy and Power received a grant from the Center for Innovation, Commercialization and Entrepreneurship Seed Fund to study precisely this issue. Specifically, VanderMeer and McDonnell will explore the possibility of harnessing high wind output during periods of low demand by dumping excess wind energy into heat, according to an ACEP newsletter. To do so, wind energy would power air source heat pumps, which would heat Alaskan homes and businesses.
VanderMeer and McDonnell’s model will utilize a demand response approach, which can account for the inherent intermittency of wind. This means that the power demand of the pumps will be adjusted to match the amount of available wind power. The pumps will be a secondary heat source; when the wind is not blowing, the building’s primary source — most commonly fuel — will provide heat.
While it is still in the developing phases, VanderMeer and McDonnell’s model has several potential benefits. First, it would draw from a renewable energy source and thus lower reliance on fossil fuels, thereby reducing greenhouse gas emissions. And, because individuals would be using less fuel, they could also save money--this is particularly important in rural Alaska, where heating costs are extremely high.
“Nome has excellent wind resources and a high cost of fuel for heat and power,” McDonnell said in an interview. “...Those factors would likely make the economics of this approach even more favorable than in Railbelt connected locations.” According to McDonnell, there are several benefits for rural communities such as Nome, including energy security and reduced exposure to fuel price volatility. There would also be less direct economic benefits--for instance, the creation of jobs for installing and maintaining the heat pumps and a potential boost to the local economy because residents are spending less on energy bills. According to McDonnell, the loads will be under direct load control, and customers in return receive a financial incentive for the services they provide to the grid.
Direct load control is when a utility remotely controls (meaning turns on and off) home appliances, according to ScienceDirect. Customers and a utility company enter into an agreement — which typically includes a financial incentive for the customer — that allows the utility company to monitor trends of usage and adjust home appliances accordingly. By reducing energy usage and eliminating energy waste, this method saves the customer money, saves electricity and reduces the demand on the system. NJUS Assistant Manager Ken Morton explained that in Nome, direct load control would mean that when there is excess wind energy NJUS would enable electric heaters in homes — in other words, exactly what the ACEP project seeks to develop.
Morton said that NJUS is currently unable to support direct load control because they lack a strong connection between meters and the control system. However, he noted, NJUS plans to replace their meters before the end of June because the technology will no longer be supported. The connection between the meters and the control system will be stronger with the new system, so NJUS would then be able to consider implementing direct load control. This is a possibility that Morton appears to be eager about, “It’s pretty cool,” he said.
McDonnell explained that much of the technology involved is already relatively advanced, so their work is primarily tailoring it to meet the needs of the specific testing. The study does not have a hard timeline, he said, because “there are a lot of moving parts,” the most important being the need to identify and collaborate with a utility.
There are also some potential difficulties for implementing the model. First, McDonnell said that the limited size of the individual market could present a challenge. However, he said, the model could be used in similar microgrid systems across rural Alaska and “the rest of the world could help overcome that limitation by making that initial investment scalable.” Thus, the hope is that the system could potentially be used elsewhere. There may also be capital cost barriers for individual households, McDonnell said. He predicts that the pump would pay for itself within five to ten years, but there would be several thousands of dollars in upfront costs--a price many households cannot afford. This deterrent, according to McDonnell, could be overcome through financing agreements or by garnering outside funding support. And the benefits could be particularly “significant” for lower income households since a higher percentage of their income goes toward heating.
