By including integrated thermal storage systems, base-load capacity factors can be achieved.
In a CSP plant, even without integrated storage, the inherent thermal mass in the collector system and spinning mass in the turbine tend to significantly reduce the impact of rapid solar transients on electrical output, and thus, lead to less impact on the grid. By including integrated thermal storage systems, base-load capacity factors can be achieved. This and the ability to dispatch power on demand during peak periods are key characteristics that have motivated regulators in the Mediterranean region, starting with Spain, to support large-scale deployment of this technology with tailored FITs.
CSP is suitable for large-scale 10- to 300-MWe plants replacing non-renewable thermal power capacity. With thermal storage or onsite thermal backup (e.g., fossil or biogas), CSP plants can also produce power at night or when irradiation is low. CSP plants can reliably deliver firm, scheduled power while the grid remains stable.
CSP plants may also be integrated with fossil fuel-fired plants such as displacing coal in a coal-fired power station or contributing to gas-fired integrated solar combined-cycle (ISCC) systems. In ISCC power plants, a solar parabolic trough field is integrated in a modern gas and steam power plant; the waste heat boiler is modified and the steam turbine is oversized to provide additional steam from a solar steam generator. Better fuel efficiency and extended operating hours make combined solar/fossil power generation much more cost-effective than separate CSP and combined-cycle plants.
However, without including thermal storage, solar steam could only be supplied for some 2,000 of the 6,000 to 8,000 combined-cycle operating hours of a plant in a year. Furthermore, because the solar steam is only feeding the combined-cycle turbine—which supplies only one-third of its power—the maximum solar share obtainable is under 10%. Nonetheless, this concept is of special interest for oil- and gas-producing sunbelt countries, where solar power technologies can be introduced to their fossil-based power market.
CSP is a proven technology at the utility scale. The longevity of components has been established over two decades, O&M aspects are understood, and there is enough operational experience to have enabled O&M cost-reduction studies not only to recommend, but also to test, those improvements.
In addition, field experience has been fed back to industry and research institutes and has led to improved components and more advanced processes. Importantly, there is now substantial experience that allows researchers and developers to better understand the limits of performance, the likely potential for cost reduction, or both. Studies have concluded that cost reductions will come from technology improvement, economies of scale and mass production. Other innovations related to power cycles and collectors are discussed below.
CSP is a technology driven largely by thermodynamics. Thus, the thermal energy conversion cycle plays a critical role in determining overall performance and cost. In general, thermodynamic cycles with higher temperatures will perform more efficiently. Of course, the solar collectors that provide the higher-temperature thermal energy to the process must be able to perform efficiently at these higher temperatures, and today, considerable R&D attention is on increasing the operating temperature of CSP systems.
Although CSP works with turbine cycles used by the fossil-fuel industry, there are opportunities to refine turbines such that they can better accommodate the duties associated with thermal cycling invoked by solar inputs.
Considerable development is taking place to optimize the linkage between solar collectors and higher-temperature thermodynamic cycles. The most commonly used power block to date is the steam turbine (Rankine cycle). The steam turbine is most efficient and most cost effective in large capacities. Present trough plants using oil as the heat transfer fluid limit steam turbine temperatures to 370°C and turbine cycle efficiencies to around 37%, leading to design-point solar-to-electric efficiencies of the order of 18% and annual average efficiency of 14%. To increase efficiency, alternatives to the use of oil as the heat transfer fluid—such as producing steam directly in the receiver or using molten salts—are being developed for troughs.
These fluids and others are already preferred for central receivers. Central receivers and dishes are capable of reaching the upper temperature limits of these fluids (around 600°C for present molten salts) for advanced steam turbine cycles, whether subcritical or supercritical, and they can also provide the temperatures needed for higher-efficiency cycles such as gas turbines (Brayton cycle) and Stirling engines. Such high-temperature cycles have the capacity to boost design-point solar-toelectricity efficiency to 35% and annual average efficiency to 25%. The penalty for dry cooling is also reduced, and at higher temperatures thermal storage is more efficient.
The collector is the single largest area for potential cost reduction in CSP plants. For CSP collectors, the objective is to lower their cost while achieving the higher optical efficiency necessary for powering higher-temperature cycles. Trough technology will benefit from continuing advances in solar-selective surfaces, and central receivers and dishes will benefit from improved receiver/absorber design that allows collection of very high solar fluxes.
Linear Fresnel is attractive in part because the inverted-cavity design can reduce some of the issues associated with the heat collection elements of troughs, although with reduced annual optical performance. Improved overall efficiency yields a corresponding decrease in the area of mirrors needed in the field, and thus, lower collector cost and lower O&M cost. Investment cost reduction is expected to come primarily from the benefits of mass production of key components that are specific to the solar industry, and from economies of scale as the fixed price associated with manufacturing tooling and installation is spread over larger and larger capacities. In addition, the benefits of ‘learning by doing’ cannot be overestimated. A more detailed assessment of future technology improvements that would benefit CSP can be found in ECOSTAR (2005), a European project report edited by the German Aerospace Center.
Concentrating solar power electricity systems are a complex technology operating in a complex resource and financial environment, so many factors affect the LCOE. A study for the World Bank (World Bank Global Environment Facility Program, 2006) suggested four phases of cost reduction for CSP technology and forecast that cost competitiveness with non-renewable fuel could be reached by 2025.
The total investment for the nine plants comprising the Solar Electric Generating Station (SEGS) in California was USD2005 1.18 billion, and construction and associated costs for the Nevada Solar One plant amounted to 245 million (USD2005, assumed 2007 base). The publicized investment costs of CSP plants are often confused when compared with other renewable sources, because varying levels of integrated thermal storage increase the investment, but also improve the annual output and capacity factor of the plant.
The two main parameters that influence the solar capacity factor of a CSP plant are the solar irradiation and the amount of storage or the availability of a gas-fired boiler as an auxiliary heater, for example, the SEGS plants in California. In case of solar-only CSP plants, the capacity factor is directly related to the available solar irradiation. With storage, the capacity factor could in theory be increased to 100%; however, this is not an economic option and trough plants are now designed for 6 to 7.5 hours of storage and a capacity factor of 36 to 41%.
Tower plants, with their higher temperatures, can charge and store molten salt more efficiently, and projects designed for up to 15 hours of storage, giving a 75% annual capacity factor, are under construction.
Because, other than the SEGS plants, new CSP plants only became operational from 2007 onwards, few actual performance data are available. For the SEGS plants, capacity factors of between 12.5 and 28% are reported. The predicted yearly average capacity factor of a number of European CSP plants in operation or close to completion of construction is given as 22 to 29% without thermal storage and 27 to 75% with thermal storage. These numbers are well in line with the capacity figures given in the IEA CSP Roadmap (IEA, 2010b) and the US Solar Vision Study (US DOE, 2011). However, the limited available performance data for the thermal storage state should be noted.
For large, state-of-the-art trough plants, current investment costs are reported as USD2005 3.82/W (without storage) to USD2005 7.65/W (with storage) depending on labour and land costs, technologies, the amount and distribution of direct-normal irradiance and, above all, the amount of storage and the size of the solar field (IEA, 2010b).
Storage increases the investment costs due to the storage itself, as well as the additional collector area needed to charge the storage. But it also improves the ability to dispatch electricity at times of peak tariffs in the market or when balancing power is needed. Thus, a strategic approach to storage can improve a project’s internal rate of return.
The IEA (2010b) estimates LCOEs for large solar troughs in 2009 to range from USD2005 0.18 to 0.27/kWh for systems with different amounts of thermal storage and for different levels of solar irradiation. This is broadly in line with the range of LCOEs derived for a system with six hours of storage at a 10% discount rate (as applied by the IEA), although the full range of values derived for different discount rates is broader. Based on the data and assumptions provided in Annex III of this report, and the methods specified in Annex II, the following two plots show the sensitivity of the LCOE of CSP plants with six hours of thermal storage with respect to investment cost and discount rates as a function of capacity factor.
The learning ratio for CSP, excluding the power block, is given as 10 ±5% by Neij (2008; IEA, 2010b). Other studies provide learning rates according to CSP components: Trieb et al. (2009b) give 10% for the solar field, 8% for storage, and 2% for the power block, whereas NEEDS (2009) and Viebahn et al. (2010) state 12% for the solar field, 12% for storage, and 5% for the power block.
Cost reductions for trough plants of the order of 30 to 40% within the next decade are considered achievable. Central-receiver technology is less commercially mature than troughs and thus presents slightly higher investment costs than troughs at the present time; however, cost reductions of 40 to 75% are predicted for central-receiver technology (IEA, 2010b).
The US DOE (2011) states its CSP goals for the USA in terms of USD/kWh, rather than USD/W, because the Solar Energy Technologies Program is designed to affect the LCOE and includes significant storage. The specific CSP goals are the following: 9 to 11 US cents2005/kWh by 2010; 6 to 8 US cents2005/kWh (with 6 hours of thermal storage) by 2015; and 5 to 6 US cents2005/kWh (with 12 to 17 hours of thermal storage) by 2020 (USD2005, assumed 2009 base). The EU is pursuing similar goals through a comprehensive RD&D program.