By 2030, scaled-up green power could meet the demands of a large grid 99.9 percent of the time, according to new research from the University of Delaware.
A mix of offshore and onshore wind, along with contributions from solar power, could provide reliable power flow during all but a handful of days in the hypothetical four-year period under study.
Moreover, researchers found that scaling up renewable generation capacity to seemingly excessive levels -- more than three times the needed load, in some instances -- proved more cost-effective than scaling up storage capacity, due to the high systems costs associated with storage technology.
"That's a lot of overbuilding," said Willett Kempton, a professor in the School of Marine Science and Policy at the University of Delaware and a co-author of the study. Much of that excess capacity would be underused during all but a few days a year, he said.
At the same time, thermal power plants face a similar problem today through inefficiency, he added.
"If you think about it, power plants burn three times the amount of fuel energy needed to produce their energy output," he said. "You burn three units of coal to get one unit of electricity."
Overgeneration would be cost effective even if all excess energy were simply dumped, according to the study. If that excess energy were harnessed -- to offset the costs of heating fuels, for example -- costs could be lowered even further.
Diversity of supply
Reliability has long been the Achilles' heel of renewable energy, which depends on intermittent weather conditions like wind and sun to generate power. However, by extending enough wind turbines and solar panels over a wide enough area, it is possible to achieve approximate reliability by shifting power from active to passive regions.
The study did not assume the introduction of new, more efficient technologies, although it did form its calculations based on 2030 technology costs and energy prices. Its models incorporated four years of energy use and weather data from within PJM Interconnection, a regional transmission organization covering about one-fifth of the United States' total electrical system.
The simulations were run on the XSEDE supercomputer network, a project of the National Science Foundation. The researchers ran 28 billion separate simulations, sifting for the lowest possible cost to achieve varying levels of reliability for 72 gigawatts of power.
The simulations found that onshore wind power was consistently the cheapest renewable option, followed by offshore wind, with solar power and limited hydrogen energy storage coming online only when the researchers asked for near-perfect reliability.
"When we modeled to cover 30 percent of the hours under study, the least-cost scenario didn't include any offshore wind or solar, Kempton said. "When we modeled for 90 percent reliability, the scenario included both offshore and onshore wind."
It wasn't until the researchers asked for a scenario in which energy supply met demand 99.9 percent of the time that solar was brought into the picture, he said.
"Solar was more expensive than wind power, but it also matched load most closely," said Cory Budischak, an instructor at Delaware Technical Community College and co-author of the study. "It's more windy at night, but sunnier during the day when you see most of your peak demand. So at 90 percent reliability you can get by with just wind, but to get that last 9.9 percent, you really need solar."
Electric car batteries as backups
By building up renewable energy capacity to around 290 percent, energy could be delivered at a low cost with very little battery storage needed, Budischak said.
"You still need battery storage, but only enough for a couple of days, rather than a couple of weeks," he said.