Patrick McDonnell, President, Agland Energy Services Inc.
April 23, 2013 | 27 Comments
Nicasio, California -- The PV industry is the subject of an increasingly confusing discussion about its future. Statements that solar PV is competitive today and will expand exponentially are countered by detractors pointing out grid limitations and studies suggesting cost shifting. What is missing from this discussion is a clear and systematic framework which provides a common vocabulary to bracket discussions about solar PV's competitiveness.
To be useful, such a framework must be applicable to most power grids around the world while recognizing that competitiveness values are specific to each grid’s mix of generation and transmission resources. Critical to the discussion must be the opportunities and challenges of integrating PV into the grid, including an honest assessment of costs and benefits. Because integration issues vary depending on penetration, discussions must be explicit about the levels under discussion.
The framework must reflect the fact that the services offered to the grid, and therefore their value, change as the level of PV capacity rises, relative to a grid’s total generation capacity. If PV’s costs continue to fall along the path that most analysts forecast, the industry’s economics will face four sequential steps based on PV’s ability to provide: Firstly, peaking power services, a premium value service from PV due to when power is available; second, intermittent wholesale power – with PV’s value based on avoiding variable fossil fuel costs; thirdly comes grid restructuring, which will require PV to be inexpensive enough to justify significant and complex grid restructuring costs; and the final phase sees the emergence of fully dispatchable electricity and ancillary services, if electricity storage becomes economical.
At each of these steps three values can be derived. First is the approximate size of the PV market. Second is the target price per kWh at which a PV array must be able to deliver electricity before it is deemed competitive at that level. Third is the capital cost of solar installations that permit delivery of electricity at, or below, that target price.
While other values, such as CO2e reductions, distributed generation benefits and solar's low environmental discharges may have greater and lesser traction in the politics of local rate-making, their inclusion or exclusion does not change the basic structure of the framework. And the target price at which PV becomes competitive with conventional generation depends on the grid. In Saudi Arabia, for example, the opportunity cost of displacing some of the nation's oil generation with PV is U.S.$10-$15/MMBTU, while in the US, natural gas is $3.50/MMBTU as delivered to a power plant. These differences and the inclusion of environmental values do not change the functional form of the framework, they only change the price. One caveat is that the first step of the framework does not apply in regions where solar's delivery of power is not coincident with peak demands.
Solar output varies with sunshine rather than in response to customer demand. Therefore no grid can rely only on solar PV without a major advance in storage technology. This constraint defines the services that solar can offer to the grid and the services the grid needs to supply independently. The “must take” character of PV requires a grid operator to manage that obligation based on low-demand days.
Over the past few years many grid managers have investigated how PV integrates into their grids at different penetration levels. This has provided a substantial base of information about how grid integration issues change as market penetration levels rise. Based on the goal of keeping the general body of ratepayers economically neutral, the net of costs/values of increasing levels of PV on the grid provide a target price against which to compare solar’s declining costs. In turn, at each price neutral point, there is a maximum level of solar penetration that works.
Phases of Solar Market Penetration
At low levels of deployment in warm regions, PV offers peak power services during daytime hours. Where air conditioning drives peak demands, PV lowers the height of the peak that must be met by conventional sources, displacing fossil fuels and a portion of the capital facilities that are required to meet those brief loads.
In this first step, PV does a good job matching load as peak demand rises, but falls off much more quickly than customer demand. This is because in the late afternoon, when air temperatures are peaking and residences are using more air conditioning, the sun has dropped to more oblique angles, diminishing solar's output. But even with this less-than-perfect match, most analysts agree that PV, at a low percentage of grid capacity, provides a high value due to real grid benefits arising from reducing the highest summer demands.
In step 2, PV is only an opportunistic supply source to the grid and imposes little or no grid integration costs on the generation or transmission resources. This means the grid realizes a value from PV to the extent that it is cheaper than the variable costs incurred by conventional systems. At this stage, solar's variable character prevents it from claiming any significant capacity value for providing reliability services.
A key element of step 2 is that there is little change to the grid’s costs arising from generation capacity, distribution or transmission facilities or the utility support services required to operate and maintain the system. Within the penetration level of step 2, any reductions in the grid’s cost of meeting peaking power services are very small and the total solar output at this step is sufficiently small to impose few significant costs for integrating the rising level of new solar installations. At this level, solar’s role is defined by its ability to reduce – but not eliminate – conventional power for selected hours of the day. Grid managers may also elect to add other values for CO2e reductions or distributed generation benefits, to the engineering values.
The transition from step 2 to step 3 occurs when PV’s installed base grows sufficiently to cause either or both of two things: base load resources become disruptively intermittent and hence more expensive per kWh; and the grid now requires the installation of expensive transmission to balance power across and between regions.
When grids must take PV whenever it is available, conventional resources must then be dispatched to fit the remaining un-met system demand, which resembles a nighttime peaking profile. Even if PV were to meet 100 percent of demand during some mid-day hours, large portions of the day still require supplies from other sources. So in step 3 there are four essential factors that define the grid's restructuring issues.