COOLING

COOLING
Utility-scale concentrating solar power (CSP) plants operate by absorbing thermal

energy from concentrated solar irradiation and converting the high-temperature
thermal energy into electrical power via a Rankine steam cycle. The Rankine
thermodynamic cycle requires the flow of energy from a hot thermal reservoir (the
solar field) to a cold thermal reservoir to generate mechanical power. The total cycle
efficiency is subject to the temperatures of the hot and cold reservoirs; a highertemperature hot reservoir or a lower-temperature cold reservoir both serve to
improve cycle efficiency and maximize power output, but the flow of some of the
heat into the cold reservoir (heat rejection) is a prerequisite to cycle operation.1
The heat rejection system can take one of several forms. Traditionally, wet cooling
has been used since it provides a low-temperature heat rejection reservoir with the
wet-bulb temperature. However, this mechanism consumes a large amount of water
via evaporation, so wet cooling is untenable in locations where the water supply is
limited. The practical alternative to wet cooling is air cooling. This configuration is
subject to the much warmer ambient dry-bulb temperature and a large temperature
rise in the air stream due to the low specific heat capacity of air. Consequently, the
negative impact on plant performance is accentuated during the hot summer
afternoon hours when both peak electricity demand and opportunity for plant
revenue are highest.
While water is the preferred medium for power-cycle cooling, its availability may be
limited by policy or cost in some locations, especially throughout much of the arid
Southwest. Under these conditions, the plant designer can opt for an air-cooled
system or, if some water is available, a hybrid design that uses both air and water
for cooling. An air-cooled condenser (ACC) condenses the steam by forcing ambient
air over a bundle of finned tubes containing the steam that exits the turbine. A
typical hybrid system includes both an ACC and a wet-cooled tower operating in
parallel, and the size of each can be adjusted depending on the design intent
A critical distinction between the effectiveness of wet- and dry-cooled systems
relates to the minimum cooling temperature each technology can achieve. Drycooled processes rely on air cooling and are limited by the ambient dry-bulb
temperature. In contrast, wet cooling processes use evaporation to reject heat and
can achieve minimum temperatures that approach the ambient wet-bulb
temperature. Wet-bulb refers to the temperature achieved by a moistened
thermometer in flowing air and reflects the reduced temperature that is possible
when evaporation from a surface is accounted for. The difference between wet-bulb
and dry-bulb temperature depends on humidity: they are equal at 100% relative
humidity, and wet-bulb is always lower than dry-bulb temperature in other
conditions. Figure 2 depicts an example of how the wet-bulb and dry-bulb
temperatures relate to the operating conditions of wet and dry condensers.
A cooling tower can easily bring water to within 5C of the wet-bulb temperature
(known as the approach temperature). This cooling water is then used to
condense the steam at a temperature typically a few degrees warmer, with the
overall difference between the wet-bulb temperature and the condensate on the
order of 10C; that is, for a wet-bulb temperature of 24C, a typical wet-cooled
condenser may operate at 34C (see Figure 2). In contrast, a dry-cooled condenser
is designed to condense steam at a temperature above the dry-bulb temperature;
this is denoted as the initial temperature difference (ITD). Because air has a low
volumetric heat capacity, the heat transferred to the air results in a large
temperature increase in the air as it passes through the ACC. (Higher air flow rates
could reduce the temperature increase, but at a high cost in fan power. Similarly,
larger ACC areas can be used, but also at a higher cost.) In the example shown in
Figure 2, the dry-bulb temperature is 32C, and even with an aggressive ITD
temperature of 16C, a dry-cooled condenser can only achieve a condenser
temperature of 48C. Wet-cooled condensers are normally designed with a closer
approach than ACCs because the marginal cost of an ACC climbs dramatically as
one nears the dry-bulb temperature. As previously shown, the efficiency of the
Rankine steam cycle depends on the condenser temperature, so as condenser
temperature climbs, power-cycle efficiency falls.
3 Cost model
3.1 Equipment cost estimation
In order to capture the impact of cooling system cost on the total installed cost of
the plant, a cost model was developed to estimate the additional expense incurred
for a hybrid plant. The air cooled and wet cooled systems were sized independently
according to the total heat rejection load and the maximum fraction of that load
required for the wet cooling system. All non-power-block plant cost values are equal
to the SAM defaults for the parabolic trough technology. The total power block cost
value is determined by adding the calculated configuration-dependent cooling
system cost to the fixed baseline cost that accounts for all of the power cycle
equipment except for the cooling system. Table 3 provides a summary of the major
cost items and selected financial assumptions used in this analysis.
The goal of the ACC cost equation is to express the condenser cost in terms of total
fin surface area. This approach accounts for scaling in total size due to heat
rejection load and size increase due to the design-point ITD. For the ACC system,
conductance (UA) per thermal load rejected is a function of the ITD. The IPSEpro
process modeling software [5] was used to develop a correlation for UA per rejected
load for an analogous air-cooled system, as shown in Eq.[5].
CUA,acc = 1310.48 T 0.793
The wet cooling system equipment costs are broken down into a wet surface
condenser cost and a cooling tower cost. The surface condenser cost is based on a
UA per rejection load value determined using IPSEpro, and the total condenser
surface area is calculated in the same method previously discussed for the ACC.
Condenser cost is calculated using a cost correlation obtained from vendor data [7].
The A0.13 condenser surface area Awc is used to determine the material cost
coefficient, where cmat = 1.222 wc . The pressure coefficient cpsi is equal to 1.05
for a full vacuum system.
497
Costwc = 1909 A WC
.cmat. cpsi
4 Results
The goal of parallel wet/dry cooling is to boost power production (and revenue) for
ACC systems during the most profitable peaking TOD periods, thereby benefiting
utilities in satisfying peak demand and plant owners in increasing production
revenue. By measuring performance with the bid price, the analysis in this paper
shows that traditional metrics can fail to produce plant configurations that are
optimized for profitability. For example, Figure 3 shows both the LCOE and bid price
plotted for a range of ACC ITDs and wet cooling fractions for a plant
Notably, the increase in wet hybridization has a markedly greater impact on bid
price than LCOE, and the maximum LCOE reduction occurs at a much lower ACC ITD
than the maximum bid price reduction. The plot of minimum bid price for each
hybrid cooling fraction in Figure 4 shows that hybrid cooling shows promise even in
cooler climates like Alamosa, CO, for heavily weighted TOD schedules. Results from
the full analysis are summarized in Table 5. Dry and hybrid cases are compared to
the baseline wet cooling system for each plant location, and optimal plant
configurations for the LCOE and bid price metrics are presented separately.
5 Conclusions
The results of this analysis show that parallel wet/dry cooling offers an opportunity
for significant reduction in water use compared to wet cooling while providing a
noticeable improvement in a bid price. Several trends in the results are of interest
and have applications in the design of heat rejection systems:
Bid price minimization offers advantages over LCOE- or production-based
approaches, and captures the potential viability of hybrid cooling for some markets.
The bid-price optimized ACC ITD for an exclusively dry cooled plant is generally
lower than for a hybrid cooling plant. The difference is most pronounced in hot
climates.

COOLING

Uploaded by

Copyright:

Available Formats

COOLING

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

COOLING

Uploaded by

Copyright:

Available Formats

COOLING

Utility-scale concentrating solar power (CSP) plants operate by absorbing thermal

You might also like