Solar Power Generation Systems

In this section, the main kinds of established solar power systems, including small-(individual) and large-scale residential power generation are classified as shown in Figure 4.2 and analyzed in terms of their overall and component performance. A typical solar-driven heat engine system for residential power (and heat) generation consists of a solar concentrating collector that drives a heat engine (e.g., a Rankine cycle). The heat engine produces shaft work at an expander that in turn drives an electrical generator; additionally, the rejected heat may serve a useful purpose (e.g., water heating). Such a system can be connected to the grid or can work independently with energy storage in various ways, as will be discussed in the next section of this chapter.

Even though they are conceptually similar, large-scale solar systems differ from small-scale systems through the fact that they use a central power plant. In large systems, a field of collectors is used to capture the solar energy, which is transmitted by means of a heat transfer fluid to a standard power plant. In small-scale systems every individual unit is equipped with a low-power heat engine, usually placed in the focal point of a solar concentrator, close to the solar receiver.

Fig 4.2 Types of solar power generation systems.

Small-scale systems are mainly based on paraboloidal dish solar concentrating collectors. As its name suggests, a solar concentrating collector's role is to concentrate the solar radiation on a small spot. This is done with the purpose of reducing the exposure of the heated surface to the environment and thus to avoid heat loses. A concentrating solar collector has two main parts, that is, a solar concentrator and a solar receiver. The solar concentrator receives solar radiation under an aperture of area Aa and focalizes the incident radiation on a small spot. The solar receiver (or absorber) is placed at the focus of the concentrator and has a small aperture area Aab. The characteristic parameter of concentrating solar collector is the concentration ratio C=Aa/Aab.

Apart from dish concentrator (of paraboloidal surface) another established option is the Fresnel mirror formed from an assembly of plane or curved surface mirrors suspended by a frame structure. The individual mirrors point toward a single focal point where the solar receiver is placed. For tracing the sun, the whole assembly rotates around the azimuth and zenith angles. A similar option is represented by Fresnel lenses which use light refraction phenomenon to focalize the incident radiation.

Based on Jaffe and Poon (1981) and Jaffe (1989) there are, discussed here, a number of small-scale designs introduced to use parabolic dish concentrators for solar thermal power systems:

• The OMNIUM-G concentrator has a 6 m diameter paneled dish which provided 7-12 kW to a Rankine engine under IT=1 kW/m2 insolation.

• The Test Bed concentrator had an 11 m paneled dish and provided 76 kW to its heat engine under the same conditions.

• Lajet designed a concentrator consisting of 24, 1.5 m diameter dishes. This system delivered 33 kW to a Brayton engine under IT=1 kW/m2 insolation.

• The Advanco concentrator provided 74 kW to its Stirling engine using a 10.6 m diameter paneled dish under the same insolation conditions.

• Again, normalized to 1 kW/m2 insolation, General Electric's Parabolic Dish Concentrator 1 used a 12 m paneled dish to provide 72.5 kW to a heat engine.

• Power Kinetics had a 9 m square-shaped paneled concentrator that delivered 28 kW to a boiler under 0.88-0.94 kW/m2 insolation.

• The Acurex Parabolic Dish Concentrator 2 used an 11 m paneled dish and was shown to have an optical efficiency of 0.88 even at concentration ratios as high as 1300.

• Boeing decided to create reflector panels and test them using the Test Bed concentrator. These panels were 0.6 x 0.7 m and provided an optical efficiency of 0.8 up to concentration ratios of 3000.

• By comparison, the ENTECH Fresnel Concentrator Lens Panel had dimensions of 0.67x1.2 m and could only provide an optical efficiency of 0.68 at a concentration ratio of 1500.

Currently, the largest single-dish power system is Australia's "Big Dish," which produces 50 kWe feeding a 500oC boiler with an aperture of 400 m2 and can operate at peak efficiencies up to 29%.

Although paraboloid dish collectors have been around as long as trough collectors, there is a cost and technological gap which needs to be closed in order to exploit their high efficiencies. By innovative design, analyzing performance-cost trade-offs and introducing technologies such as direct steam generation or organic Rankine cycles, the gap can be tightened allowing for low-cost, high-efficiency, large-scale, and residential applications to be viable.

The cost of a solar dish power generator is tightly correlated to the optical performance of the solar concentrator. The performance factors of a dish system can be greatly degraded with changes in geometry, and therefore, accuracy and rigidity are important for their design.

Based on Jaffe and Poon (1981) one can extract that for an optical efficiency of 0.90-0.93, concentration ratio of 2000-5000, intercept factor of 0.98, and a lifetime of 30 years, a low-cost price estimate is $200-350/m2. Note that all cost estimates in this chapter are reported for 2008 monetary value.

Another estimate can be made based on the Acurex concentrator - described by Overly and Bedard (1981) - with optical efficiency of 0.86 and concentration ratio of 1900. This estimate of $330/m2 involves the assumption of a large dish system with production quantities of 100,000 units per year.

Back silvered glass is standard for the mirror component of the design and have about 94% reflectivity. The reflector can be a single layer which is more efficient and more expensive, or it can be broken into components, which is cheaper but less efficient. Another interesting option is the stretched membrane mirror. Manufacturing stretched membrane mirrors involves a vacuum process, as well as non-uniform loading. Stretched membrane technology was initially used to create highly effective, low-cost heliostats. Singular element stretched membrane mirrors have demonstrated optical efficiencies of 0.915. Although the performance of stretched membrane mirrors is lower than that of the best glass-metal mirrors, stretched membrane mirrors cost much less. Stretched membrane mirrors are often made in facets, in which, several smaller elements are connected to form a larger element. The smaller elements are generally produced in sizes of 3-4 m diameter, which is the approximate size of the projected residential solar dish unit presented in the case study below (see, e.g., Alpert and Houser, 1988).

The dish support can be a solid metal structure or a truss structure which may include tension cables for additional support. The receiver support can be central or extended structures; however, shading and distortion influence the effectiveness of the support. The foundation is often a concrete ring for azimuth rotation.

Two methods of tracking control are used where in the first, sensors provide optical feedback to allow for variable tracking, and in the second, the system is pre-programmed to follow the sun. For non-reactive tracking, there are two more options whereby polar tracking allows a single axis motion over the course of a day and adjusts a second axis daily or weekly, and azimuth tracking allows for constant two-axis tacking (see Jaffe, 1983a). Hydraulic-drive tracking systems are simpler while electric-drive tracking systems are cheaper.

For low-power generation systems specific to solar dish systems, Rankine cycles operated with organic fluids (e.g., toluene), known also as Organic Rankine Cycles (ORC), are believed to be the most effective. This fact is due to the gas dynamic characteristics of organic fluids that are suitable for development of cost- effective turboexpanders.

Steam cannot be used in low-power applications (below MW) with regular steam turbines. To use steam in low-power Rankine cycles, a special steam expander (e.g., reciprocating or pulse turbines) must be considered. Ammonia water is another working fluid worth being considered. It should be mentioned that if ammonia, ammonia water, or organic fluids are used in the Rankine cycle, then low-cost refrigeration compressors of scroll or screw type can be used as expanders. This feature increases the marketability of independent low-power solar-driven generators. An example of ammonia-water Rankine generator that is able to match the temperature profiles at both source and sink has been developed by Zamfirecu and Dincer (2008a,b). Due to the feature of ammonia-water solution to vary its temperature at vapour-liquid phase change, the temperature differences at sink and source can be minimized for better energy and exergy efficiency.

Solar-driven Rankine cycle systems compete with other solar thermal energy conversion alternatives, each of them having notable drawbacks and advantages. A review by Kongtragool and Wongwises (2003) shows that the solar power systems based on Stirling engines operate at very high pressures, of the order of 200 bar and temperatures in the range of 700-800oC working with helium or hydrogen. These gases leak easily, which raises maintenance problems. Additionally, hydrogen is highly flammable which imposes severe safety issues; however, Stirling systems are very compact and reach high efficiency around 40% (of the engine). As reported by Kongtragool and Wongwises (2003) an efficiency of 22% was obtained for a dish-Stirling system operating for 10 h/day. Another system, known as SAIC/STM SunDish obtained a peak efficiency of 26% for 23 kW. One of the main drawbacks of using Stirling engines in solar applications is related to the long warm-up time needed, which is in contradiction with the reality of solar energy's fluctuating nature.

Open-air Brayton cycle engines mounted at the dish focal point were also used in some applications (see e.g., Jaffe, 1983a). They operate efficiently at higher receiver temperatures than usual for Stirling and Rankine cycles, that is, over 1000oC where Brayton engines may attain over 26% efficiency. At lower temperatures, the system efficiency drops under 20%.

Modern large-scale solar technology came about in the 1980s when nine substantial power generating stations totaling 354 MW power generating capacity were built in California's Mojave desert. These systems are based on parabolic trough collectors which are considered to be the most proven of all solar technologies. These plants make use of linear parabolic reflectors that concentrate the solar energy on lengths of tubing. Within the tubing, heat is collected and transferred to steam, which is then passed through a steam Rankine cycle. In the summer months, trough plants can operate from 10 to12 h/day solely on solar energy.

Solar

Expansion Solar Preheater Thermal Vessel Reheater Energy Storage (optional)

Fig 4.3 Schematic of 30 MW hybrid solar trough power plant (modified from Kearney and Miller, 1988).

Solar

Expansion Solar Preheater Thermal Vessel Reheater Energy Storage (optional)

Fig 4.3 Schematic of 30 MW hybrid solar trough power plant (modified from Kearney and Miller, 1988).

To date, most of the large-scale solar plants are hybridized with a fossil fuel (gas/coal) burning system to keep the flow of power relatively constant. Notable examples of solar power plant technology obtain a Power Output per Unit of Receiver Area ranging from 120 to 175 (W/m2) and an average operating temperature of 360oC. The Luz LS-3 is regarded as the most advanced trough technology presently available. The performance factors of several systems are presented in Table 4.1.

A significant example of large-scale solar trough power plant is illustrated in Fig. 4.3 and refers to the 30 MW plant VI referenced in Kearney and Miller (1988) which serves as a benchmark for solar technology with a net efficiency of 10.7% and a capital cost of $150 million. The plant is hybridized with 25% of its energy coming from natural gas in low solar radiation periods. By comparison, an 80 MW Integrated Solar Combined Cycle System with the constant addition of fossil fuel assistance would cost $3850/kW, with an efficiency of 13.5%. In the future, with advancing technology, economy of scale, and by adding thermal storage capacity, tilted collectors, and direct steam generation, efficiencies are predicted to increase by 30%, and cost reductions of 30%/kW are predicted by 2030. Additionally, operation and maintenance costs are expected to decrease by 25%.

The second representative large-scale power plant is based on solar tower technology. The solar tower is in fact a large size Fresnel lens assembly that focus the solar radiation on a central receiver placed on a tower at certain height. Being of large size, these systems are suitable for using thermal storage of solar radiation in molten salts (see Ortega et al., 2008).

Table 4.1 Solar collector characteristics.

Acurex

M.A.N.

Luz

Luz

Luz

Luz

Collector

3001

M480

LS-1

LS-2

LS-2

LS-3

Year

1981

1984

1984

1985

1988

1989

Area (m2)

34

80

128

235

235

545

Aperture (m)

1.8

2.4

2.5

5

5

5.7

Length (m)

20

38

50

48

48

99

Collector diameter (m)

0.051

0.058

0.042

0.07

0.07

0.07

Concentration ratio

36:1

41:1

61:1

71:1

71:1

82:1

Optical efficiency

0.77

0.77

0.734

0.737

0.764

0.8

Receiver absorptivity

0.96

0.96

0.94

0.94

0.99

0.96

Mirror reflectivity

0.93

0.93

0.94

0.94

0.94

0.94

Receiver emittance

0.27

0.17

0.3

0.24

0.19

0.19

@Temperature, oC

300

300

350

350

Operating temperature, oC

295

307

307

349

390

390

Source: Compiled from Mackay and Probert (1998), Winter et al. (1990).

Source: Compiled from Mackay and Probert (1998), Winter et al. (1990).

Luzzi and Lovegrove (1997) opined that among all solar energy conversion systems, solar dish fields with centralized power generation are believed to be the most viable solution for the future and provide the highest efficiency of all. Australian National University consecrated large efforts in the development of solar dish heat engines. They demonstrated a 28 dish field system having the peak solar power of 2 MWe and supplemented with 4 MWe gas-fired plant. Steam at 50 MPa and 500oC is generated by each solar dish for an equivalent electrical power of 50 kW corresponding to an insolation of 950 W/m2 (see Kaneff, 1999).

Table 4.2 Summary of performance of large-scale systems.

Technology

Parabolic trough + oil

Solar tower system

Solar dish field

Mean net efficiency

14

13.8

19

Specific power generation (kWh/m2 - yr)

308

316.5

340

Levelized capital costs ($/kWh - yr)

2.39

4.22

2-4

Operation and maintenance (c$/kWh)

4.96

6.05

4-6

Levelized electricity cost ($/kWh,)

0.248-0.295

0.24-0.31

0.2-0.4

Source: Compiled from Ortega et al. (2008) and Lovegrove et al. (2007).

Source: Compiled from Ortega et al. (2008) and Lovegrove et al. (2007).

Another option, proposed by Luzzi and Lovegrove (1996) at the Australian National University, consists of using ammonia as an energy transfer and chemical energy storage medium. In this case, the receiver of each solar dish unit has a chemical reactor for ammonia decomposition. This reaction is endothermic, and the reverse reaction, ammonia synthesis is exothermic according to NH3 + 66.5kJ/mol o 0.5N2 + 1.5H2.

The decomposition products, hydrogen and nitrogen, are stored under pressure in a specially devised vessel. For power generation, the hydrogen and nitrogen are combined in a synthesis reactor and deliver the associated heat of formation. Two options were proposed for power generation. One involves driving a typical Rankine cycle with the reaction heat. The other involves using a Brayton cycle, where the produced ammonia is expanded for power generation. Two other thermochemical storage options were considered: the sulfur trioxide 2SO3 + 196.4kJ/mol o 2SO2 + O2 and the ammonium hydrogen sulfate NH4HSO4 + 132kJ/mol o NH3 + H2SO4

20 MWe

Fig 4.4 Levelized electricity cost (LEC) for large-scale solar systems; compiled based on data from Kaneff (1999) and Price (2003).

Solar generators imply very reduced maintenance costs. Thus, the cost of generated electricity is levelized based on the investment cost, maintenance cost, and lifetime of the system. The following formula applies for calculating the level-ized electricity cost

In Table 4.2, we present a comparison between the three kinds of large-scale solar power generation systems discussed above in terms of efficiency and costs. An additional comparison is presented in Fig. 4.4 where, based on data from Kaneff (1999) and Price (2003), we correlated the levelized energy cost (LEC) of solar power systems for paraboloidal dish fields and for parabolic trough systems. In the case of dish systems two scenarios were considered, namely systems relying 100% on solar energy and hybrid systems that use 50% fossil energy. Dish-based systems are more cost competitive than parabolic trough, mainly due to their better performance. As expected, the LEC decreased with the installed performance. It is worth noting that the same trend is expected for individual dish system units, namely their associated LEC decreases with production size.

One important aspect involves the CO2 mitigation expected from the foreseen solar energy expansion. Based on data from Brackmann (2008), which predicted the trend of expansion of solar energy utilization in future years, we obtained the plots presented in Fig. 4.5. Figure 4.5a shows the estimated CO2 saved per year from using solar energy power generation instead of fossil fuel. Figure 4.5b shows the predicted investment evolution from present to the year 2025.

70000

c

£

60000

m

50000

ai

>-

40000

<u

ti ■a

S

20000

V)

10000

o

o

2005 2010 2015 2020 Time (Year)

2025

18000 15000 -12000 -9000 -6000 -3000 -0

2005 2010 2015 2020 Time (Year)

18000 15000 -12000 -9000 -6000 -3000 -0

2005 2010 2015 2020 Time (Year)

2025

2005 2010 2015 2020 Time (Year)

Getting Started With Solar

Getting Started With Solar

Do we really want the one thing that gives us its resources unconditionally to suffer even more than it is suffering now? Nature, is a part of our being from the earliest human days. We respect Nature and it gives us its bounty, but in the recent past greedy money hungry corporations have made us all so destructive, so wasteful.

Get My Free Ebook


Post a comment