New Ideas to Create Sun Energy and New Business

There have been two ways to harness the sun to generate electricity: build a large, single array that might produce as much as 50 megawatts to feed directly into the grid in much the way the output of a traditional power plant would, or, install small rooftop units and use net metering to offset a homeowner's electric bill by the amount the rooftop solar panels contribute. From the utility's perspective, the rooftop option is really a net loss. And with a payback period of perhaps six or seven years, although that's much shorter than the 12 to 15 years for rooftop solar  installed just a few years ago, it remains a rather low-return, long-term investment for the homeowner.

But there may be a third option that, at least theoretically, could be attractive to both utilities and power customers: solar aggregation, treating several hundred or more individual rooftop installations as a single, virtual distributed generation power plant. It's an idea being actively pursued by some of the larger PV manufacturers and being considered by a small number of utilities. In many ways, aggregation looks attractive. Depending on how the business model is developed, it could offset the cost of installation, help meet increasingly demanding state requirements for renewable power and help utilities take advantage of grid capacity in locations where demand is increasing. At least, that's how the theory goes.

But as with so many virtual versions of physical realities, the question remains: Can a virtual distributed power plant meet the needs of real-life power customers and the demands of a real operating business?

Tom Werner, CEO of SunPower, a leading manufacturer of solar panels, with the largest  installed base in the United States, believes aggregating individual rooftop installations is a viable way for utilities to capitalize on distributed generation. One model he suggests would have a utility own a number of rooftop installations distributed across a radius of a couple of miles or more. The utility would compensate a homeowner either by leasing the roof or by offering especially attractive financing to install a system. "It's a good way for utilities to quickly increase capacity nearer demand," Werner says. The utility would in essence be building a virtual small power plant and thus offset the need to add large-scale capacity.

Two hurdles facing wide adoption of Werner's idea are the inherent variability of any renewable source and the costs of connecting to and managing individual units on the grid. He believes aggregation could address these concerns. If the assets - rooftop panel arrays - are spread over a geographic range, perhaps as much as 25 miles, it is possible to reduce variability caused by intermittent clouds, for instance, so that it approximates demand-side variability. And the sheer economies of scale realized by installing and maintaining multiple units within a reasonable distance can cut the costs of integration with the grid.

But cost remains a significant obstacle. Wide deployment of aggregated units meant to supplement generating capacity would require heavy investment in so-called smart inverters, for instance, inverters that go far beyond the simple need to convert DC current to AC. Eran Mahrer, executive vice president for strategy and research at the Solar Electric Power Association, points out that "much wider deployment of smart inverters would be essential if a virtual plant formed by aggregating individual solar units were to provide `edge of- grid' capabilities."

Mahrer notes that while there are locations, mainly in the Southwest, where solar in general is cost-competitive with other fuels, outside of those areas and in jurisdictions where utilities do not enjoy a very low-cost structure, solar still operates at a cost premium. As for aggregating  individual units, "even with all the benefits, it is challenging to make an economic argument for distributed resources over central stations. And that includes distributed, aggregated solar, versus a single, utility-scale solar installation, not simply solar versus natural gas. The costs are coming down, and the argument might well change over the next three to eight years," he says.

Another hurdle Mahrer sees is the need to develop and deploy much more sophisticated  monitoring systems to track and control the instantaneous operation of a dispersed, virtual power plant. "There are issues with telemetry, communicating back and forth with thousands of systems, characterizing them as a single resource for the benefit of the utility. And it's not clear exactly who should be responsible for such a system and who should be compensated for it," he says. Nor, for that matter, is it clear what the architecture of a system that involved distributed-generation, virtual solar plants would look like. Would it be owned and operated by the utility or by a large solar manufacturer, such as SunPower or Solar City? Just how would customers be compensated for use of their roofs? How would the costs and benefits of advanced telemetry systems be allocated?

Most of the challenges to the deployment of aggregated virtual solar plants are typical of the challenges faced by any new concept, whether technical or structural. The idea of aggregated solar is very new, and while there are a handful of projects that are looking at it, they are very small aggregations, according to SEPA's Mahrer, on the order of 1megawatt. "That is what I'd call the state of the art today. They have yet to be scaled up."

That, however, may soon change. The Los Angeles Department of Water and Power has solicited bids for a combination of a relatively large centralized solar generation facility that will be paired with a 10-megawatt aggregated system in the city of Los Angeles. Michael Webster, LADWP's assistant director for power systems planning and development, says it's part of the utility's feed-in tariff program. "We've taken one of our large pieces of property out in the desert and said, we'll let you build 50 megawatts out there, but the price for entry to do that is you also have to offer 10 megawatts by aggregating local rooftops. They would be responsible for finding those sites. They could target Walmart or other large warehouses or apartment buildings or individual homes, although I can't imagine residential rooftops would be cost-effective. They would get a set price for the 50 megawatt site and a different price for the distributed sites, and it's that price that they will be bidding on," he said. Although Webster wouldn't say who the bidders are, SunPower's Tom Werner confirmed that his firm is participating.

The cost of solar, whether aggregated or not, has declined rapidly over the past few years, but like any interruptible source, it is difficult for utilities to count on it for capacity. "We get a little capacity benefit out of solar, but not a lot," says LADWP's Webster. "You still need to be able to push a button and either turn on a power plant or push a button and turn off your load." The game-changer, of course would be high-efficiency, low-cost storage. But like aggregated, virtual solar plants, the big cheap battery is still somewhere in the future.

To subscribe or visit go to:  http://www.riskcenter.com

http://riskcenter.com/articles/story/view_story?story=99916042