The future of solar energy: concentrated solar power vs. photovoltaic power plants
The “solar power tower” is a catchy name for (one of the options) of harvesting solar energy today. This technology uses computer-controlled mirrors spread over thousands of hectares to reflect sun beams towards the solar power station where water is heated to high temperatures to create steam and produce electricity with a generator. While most people embrace the idea of using the sun to produce electricity, this particular method is starting to receive some scrutiny. A recent article in The Wall Street Journal questions the “avian fatalities,” which occur when birds fly through the exceptionally hot rays aimed at the tower and catch fire in midair. Environmental activists are starting to act to save the birds and the bureaucratic machine of permits and regulations has started up.
This article made me think about two issues in solar energy: the development of Concentrated Solar Power (CSP) plants and Photovoltaic (PV) Power plants, and whether centralized solar power plants will have a larger share than distributed solar systems.
CPS versus PV
When discussing CSP one should consider more than just the power towers, as there are only a few that are in operation. More common are parabolic trough solar thermal power plants, which are less harmful to birds. Parabolic trough systems are often applied in hybrid systems paired with conventional power plants. The conventional part of the plant provides power throughout the night, and the solar energy is added to the total capacity during the day.
But when it comes to producing electricity from the sun, solar PV panels are also contributing in PV power plants. The debate of whether CSP or PV power plants will prevail has been argued for several years. When looking at current and future price levels CSP has—and will have—the highest levelized cost of electricity (LCOE; €/kWh). Due to large price reductions in PV over the last few years the LCOE of PV is about half the cost of CSP, and will remain so until 2030 .
Unlike PV—besides pricing—CSP faces many other challenges focused around water for cooling CSP; the speed at which a PV plant can be built compared to CSP; and PV’s proven technology. When it comes to financing, these factors may push investors more towards PV than to CSP.
However, one of the key benefits of choosing CSP over PV is be that CSP plants can more easily provide ancillary services and provide dispatchable power on-demand using long-term storage. Combining these features in a hybrid power plant could make CSP competitive with PV in the future.
Centralized power plants or distributed systems?
The Wall Street Journal article claims that “distributed solar energy is better” given that the same energy would be provided without environmental damage. I would like to stress that it’s important to compare the same technologies. The bird fatalities are caused by a power tower, and distributed solar energy is considering PV. Indeed, PV will not cause bird fatalities as often, but still the comparison between large scale (PV) power plants and distributed PV is interesting.
Of course economy of scale is an advantage for PV power plants. Globally, multi-megawatt PV power plants are realized for less than 1 euro per watt-peak. A top-50 list of the largest PV power plants ranging from 50 MW to more than 300 MW can be found here. For private households the investment price can be a factor 1.5 higher, but capacity numbers in this sector have also been growing exponentially recently.
On the other hand, PV power plants without storage—like CSP power plants without storage—face the challenge of feeding lots of power to the grid at the time it’s produced. At that scale it has to compete in the electricity market where energy prices are often lower than the production price. Individual households can rely on self-consumption and possibly subsidies in order to get enough remuneration for their produced electricity, but utilities often cannot.
In a recent grid study (in Dutch) of PV potential in the Netherlands, DNV GL investigated the effects on the electricity grid of large PV penetrations. The study examines how much distributed PV the current infrastructure can handle to equal the potential of a central (solar) power plants. Currently, 1 GWp of PV is installed throughout the Netherlands in a homogeneous distribution. When producing at peak production, this capacity can be compared to one conventional coal or gas power plant of 800 MW. The study shows that there is an abundance of roof area which could allow the PV capacity to grow to tens of GWs. Using DNV GL’s Smart Grid Profile Model we calculated that the low voltage grid is the limiting factor for PV penetration. This capacity is mainly constrained by local transformers, but we showed that with a homogeneous distribution of distributed PV in the Netherlands can grow up to 16 GW without any new measures. Putting smart grid measures in consideration can easily increase this capacity to several tens of GW. If the distribution of PV is not evenly distributed, network integration issues will appear earlier and these should be addressed case by case.
To conclude, distributed PV can easily reach the size of large scale power plants, but the optimal solution isn’t focused on having as much PV as possible. A clever mix of renewable energy sources are needed—amongst others factors—including solar, wind, and biomass energy. Energy storage will be necessary to overcome remaining issues with intermittency and seasonal effects of renewable energy production.
 Franuhofer Institute for Solar Energy Systems ISE. (November 2013). Levelized cost of electricity renewable energy technologies.