Amidst the political battle over whether wind andsolarpower is destabilizing the nation’s electric grid, it’s important to remember that utility-scale renewables can also be resources for grid stability -- as long as the technical, regulatory and economic issues can be hammered out.
Over the past year or so, First Solar, California grid operator CAISO, and the National Renewable Energy Laboratory have been testing out how a 300-megawatt solar farm in the Mojave Desert can provide grid frequency regulation, voltage and reactive power control, and other valuable grid-balancing tasks.
While the 300-megawatt solar farm isn’t storing and shifting its energy to different hours of the day, it can run at lower than full capacity to provide headroom to shift up or down along with CAISO wholesale market prices. Its inverters can also manage the voltage fluctuations that may arise from intermittency of its output from passing clouds, and meet CAISO’s frequency regulation signals at a speed that exceeds natural-gas-fired power plants.
These make up a significant stack of grid benefits, if they can be made part of everyday operations. Late last month, Clyde Loutan of CAISO, Mahesh Morjaria of First Solar, and Vahan Gevorgian of NREL hosted a webinar to discuss the latest technical paper (PDF) on the project, taking questions on subjects ranging from inverter operations to practical market applications, all centered on that move from concept to reality.
The hosts noted that First Solar’s implementation builds on the core grid support that's a key component of large-scale solar projects today. The North American Electric Reliability Corporation has established some core requirements for voltage and frequency ride-through capabilities to avoid tripping offline prematurely, for example, and to provide some level of reactive support. That includes the ability to support up to 0.95 power factor, leading or lagging at the high side of the collector step-up transformer.
These kinds of requirements are being built into large interconnector agreements today, giving most solar farms the basic tools to interact with the grid.
Likewise, today’s utility-scale solar inverters are capable of providing power and reactive-power controls, and they also have switching frequency capability for speedy response. That’s in contrast to when First Solar began this kind of research about seven years ago, when it had to work on special high-speed communications with inverter makers.
For frequency regulation, this puts First Solar’s unit in a different class than natural-gas-fired generators, and more akin to batteries and other fast-responding assets. CAISO is already looking at First Solar’s inverter test results to incorporate into its future ancillary services market design.
It’s harder to determine the value of deliberately holding back a certain amount of solar generation -- say, 10 percent or so -- in order to be able to move production up and down in line with grid operator market signals. While utility-scale plants can actually respond much faster than conventional generation, deciding if it’s worth doing so will depend on whether the revenues available in ancillary services markets can make up for lost kilowatt-hour sales.
Of course, this control depends on the solar resources available -- if a massive cloud bank moves overhead, there’s not much that First Solar can do about it. But as long as it doesn’t happen all at once, leaving that headroom can also give it the flexibility to ramp down more gradually in order to minimize grid disruption.
First Solar also addressed concerns about the impact on solar modules of holding back generation. Because only a small fraction of energy falling on the modules is being converted into usable power anyway, turning them off completely, or placing them in an open-circuit condition, shouldn’t lead to any problems beyond elevated temperature.
But if a solar farm is being curtailed -- an increasingly likely option in solar-rich markets like California -- then it makes more sense to use the plant in a different way. Beyond that, it’s an important step in reducing the reliance on conventional generation plants as the sole source of these services during times of curtailment, when solar power is otherwise unable to help.
First Solar’s inverters can even offer frequency and voltage support at night, since the inverter can essentially use grid power, change its phase angle, and send it back to the grid as reactive power -- minus the losses due to the inverter’s AC-to-DC-to-AC conversion process. In this sense, it works like a STATCOM device, which are an increasingly common part of wind farm developments in grid environments that require additional stability at the point of interconnection.
These points are timely ones, given that the Trump administration, under Energy Secretary Rick Perry, has ordered a study to determine if utility-scale renewables are forcing the premature retirement of baseload power plants and affecting long-term grid reliability. While the study has been framed as a broad look at wholesale market structures, federal policies, energy mandates and tax subsidies, many believe it is a political gambit designed to label distributed resources as destabilizing forces on the grid.
NREL and First Solar are working on future plans to demonstrate the “synthetic inertia” capability of utility-scale PV plants. Large spinning power plants provide inertia to grid electricity that helps stabilize it across many different parameters, while solar and wind power use inverters that lack this quality. But the latest developments in power electronics and software controls are allowing inverters to be orchestrated in a way that mimics this stability, helping to overcome some of the longer-range renewable integration challenges facing the market.