Scholarly article on topic 'Assessment of Improved Molten Salt Solar Tower Plants'

Assessment of Improved Molten Salt Solar Tower Plants Academic research paper on "Earth and related environmental sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Energy Procedia
Keywords
{"concentrating solar power" / "economic assessment" / "direct absorption" / cavity / "liquid film" / "temperature increase" / "supercritical steam cycle"}

Abstract of research paper on Earth and related environmental sciences, author of scientific article — Cs. Singer, S. Giuliano, R. Buck

Abstract The temperature level increase of molten salt solar tower plants is one important task on the development schedule to increase the system's overall efficiency. In the conventional power plant technology, modern supercritical steam power plants work with life steam temperatures near 620°C. To apply these modern power blocks in molten salt solar tower plants, salt temperatures near 650°C are required. Today's molten salt tower plants reach salt temperatures of 565°C. To follow the positive development tendencies of fossil power plants, in the present work the combination of supercritical steam power plants with solar towers is analyzed. As stated in recent literature (Kolb, Kelly, Singer), the state of the art tubular central receiver concept coupled with increased steam parameters of the power block (300bar/600°C/610°C) shows a marginal potential to decrease LEC. The identification of future concept innovations with significant cost reduction potential in the field of molten salt tower plants, the enhancement of the receiver efficiency as well as the assessment of critical aspects related to their feasibility is the task of this paper. Therefore this work first focuses on the state of the art molten salt tower technology and related improved concepts with molten salt. The selected improvements follow on the one hand the aim to increase the receiver's maximum temperature level to approximately 650°C to be able to feed modern supercritical power blocks, on the other hand the aim to require a low development effort. After the reference concept is characterized, the objective is the systematic comparison of receiver technologies, which show a potential of improved thermal efficiencies. The assessed receiver candidates are tubular and direct absorption receivers, either external or located in a cavity. The analyzed power level is 125 MWel, while the steam process parameters are varied from low temperature level (550°C) and subcritical steam to high temperature level (620°C) and supercritical steam. With this, selected molten salt power plants were specified and then modeled with different sizes of solar fields and different storage capacities and analyzed on an annual basis. The results show, that the assessed concept options close to the state of the art require drastically decreased investment costs to end up with significant LEC reduction and cost-competitiveness. The increase in overall efficiency of these improved concepts is compensated by their higher financial effort. Results of further concept assessments as well as the sensitivity analysis of parameters and costs are described.

Academic research paper on topic "Assessment of Improved Molten Salt Solar Tower Plants"

Procedía

Assessment of improved molten salt solar tower plants

Cs. Singer, S. Giuliano, R. Buck

Solar Research, German Aerospace Center (DLR), Pfaffenwaldring 38-40, 70569 Stuttgart, Germany

Abstract

The temperature level increase of molten salt solar tower plants is one important task on the development schedule to increase the system's overall efficiency. In the conventional power plant technology, modern supercritical steam power plants work with life steam temperatures near 620°C. To apply these modern power blocks in molten salt solar tower plants, salt temperatures near 650°C are required. Today's molten salt tower plants reach salt temperatures of 565°C. To follow the positive development tendencies of fossil power plants, in the present work the combination of supercritical steam power plants with solar towers is analyzed. As stated in recent literature (Kolb, Kelly, Singer), the state of the art tubular central receiver concept coupled with increased steam parameters of the power block (300 bar / 600°C / 610°C) shows a marginal potential to decrease LEC. The identification of future concept innovations with significant cost reduction potential in the field of molten salt tower plants, the enhancement of the receiver efficiency as well as the assessment of critical aspects related to their feasibility is the task of this paper.

Therefore this work first focuses on the state of the art molten salt tower technology and related improved concepts with molten salt. The selected improvements follow on the one hand the aim to increase the receiver's maximum tempera ture level to approximately 650°C to be able to feed modern supercritical power blocks, on the other hand the aim to require a low development effort. After the reference concept is characterized, the objective is the systematic comparison of receiver technologies, which show a potential of improved thermal efficiencies.

The assessed receiver candidates are tubular and direct absorption receivers, either external or located in a cavity. The analyzed power level is 125 MWel, while the steam process parameters are varied from low temperature level (550°C) and subcritical steam to high temperature level (620°C) and supercritical steam. With this, selected molten salt power plants were specified and then modeled with different sizes of solar fields and different storage capacities and analyzed on an annual basis. The results show, that the assessed concept options close to the state of the art require drastically decreased investment costs to end up with significant LEC reduction and cost-competitiveness. The increase in overall efficiency of these improved concepts is compensated by their higher financial effort. Results of further concept assessments as well as the sensitivity analysis of parameters and costs are described.

© 2013 Cs. Singer. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

SelectionandpeerreviewbythescientificconferencecommitteeofSolarPACES2013underresponsibilityof PSEAG. Final manuscript published as received without editorial corrections.

Keywords: concentrating solar power; economic assessment; direct absorption; cavity; liquid film; temperature increase; supercritical steam cycle

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Energy Procedia 49 (2014) 1553 - 1562

SolarPACES 2013

1876-6102 © 2013 Cs. Singer. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/Kcenses/by-nc-nd/3.0/).

Selection and peer review by the scientific conference committee of SolarPACES 2013 under responsibility of PSE AG. Final manuscript published as received without editorial corrections. doi: 10.1016/j.egypro.2014.03.164

1. Introduction

In the past the fossil fired steam power plant park developed from low over marginal to high steam process parameters. Modern steam turbine processes have today life steam parameters of around 300 bar / 620°C / 610°C, as with higher temperatures and pressures higher efficiencies of the steam turbine process can be reached. That leads to lower fuel consumption and lower levelized electricity costs (LEC) and also to lower greenhouse gas emissions. These development tendencies could be pathbreaking also for today's development and market introduction of point focusing solar thermal tower plants (STTP) with central receiver system (CRS) technology.

For this, the question needs to be answered, if higher process parameters can support the learning curve of the STTP technology to reach lower LEC and to get earlier cost competitive compared to the fossil alternatives. Latest publications describe theoretic studies related to this topic. The predicted LEC (0.096 US$/kWhel) of a 565°C (receiver outlet temperature) subcritical baseline plant (Rankine cycle parameter ^ 125°bar°/°540°C°/°540°C) was compared by Kolb [1] with possible future-generation plants that operate at 600°C receiver outlet temperature (300 bar / 591 C / 590 C) or at 650°C (330 bar / 630°C / 630°C). The conclusions of his work predict that ~8 % reduction in LEC (reduction to 0.09 US$/kWhel) can be expected by raising the salt temperature to 650°C. He states also, that most of that benefit (~5 %) can be achieved by raising the temperature to only 600°C (0.091 US$/kWhel). Kelly [2] determined if supercritical heat transport fluids of the CRS in combination with ceramic thermocline storage systems offer a reduction in LEC compared to the baseline state of the art nitrate salt concept. In his study the baseline concept uses a nitrate salt receiver (566°C receiver outlet temperature), two-tank (hot and cold) nitrate salt thermal storage, and a subcritical Rankine cycle (125°bar°/°540°C°/°540°C). His results predict equal ratings between the reference (0.167 US$/kWhel) and the innovation with increased receiver outlet temperatures feeding a supercritical steam cycle with 300°bar°/°591°C°/°590°C steam parameters (0.168 US$/kWhel). Singer et al. [3,4,5] first assessed on a less detailed level STTP driven ultrasupercritical steam cycles (USC, 350 bar / 700°C / 720°C ) fed with heat by tubular CRS using varied heat transfer media (HTM). The conclusions predict a LEC reduction potential of ~10 % if a high temperature alkali-chloride salt is used as the HTM and up to ~15 % in the case of highly effective HTM coolants like liquid metals. In a later assessment the economic potential of innovative receiver concepts with different solar field configurations for SC steam cycles (300 bar / 600°C / 610°C) were analyzed. One of the conclusions of this study was, that a tubular CRS with increased receiver outlet temperature of 620°C can lead to ~5 % LEC reduction compared to the state of the art reference concept with same parameters like Kelly and Kolb assumed. Another conclusion was that direct absorbing cavity receivers with liquid film cooling could lead to an LEC reduction of up to ~7 %. After the optimization of the internal direct absorbing receiver concept (IDAR) with liquid film cooling the results showed a possible LEC reduction of up to 8.4 %. In both of the last works it was concluded also, that with the IDAR at 565 °C receiver outlet temperature and a subcritical baseline power block a potential to decrease the reference LEC by ~4 % to ~7 % exists. For all assessments different power levels were chosen from the authors. Kelly worked with a power level of 400 MWel and a multi tower system with two towers and two fields and without optical interaction of these two modules. Kolb's gross plant power rating varies from 139 MWel to 167 MWel, whereas a single field tower configuration is used. Singer et al. worked at the beginning with a USC power level of 50 MWel to be able to compare their first results to the ECOSTAR study [6]. For the later assessments the power level was raised to 200 MWel, whereas a multi tower configuration of four, later seven tower-field modules without optical interaction was assumed. The estimated component specific investment costs of all studies arise from literature or come from accomplished companies located in the energy sector.

With the focus on increased receiver temperatures, in this paper three different receiver concepts are compared:

• state of the art tubular receiver,

• cylindrical, face down cavity receiver with internal absorber tubes and

• cylindrical, face down cavity receiver with inclined absorber walls cooled by a molten salt film.

The receiver concepts are compared to each other applying a single tower configuration and an overall power level of 125 MWel. For this, subcritical and supercritical power blocks with two different temperature levels are assumed:

• 162 bar / 550°C / 550°C (subcritical, 565°C receiver outlet temperature)

• 162 bar / 620°C / 620°C (subcritical, 635°C receiver outlet temperature)

• 250 bar / 550°C / 550°C (supercritical, 565°C receiver outlet temperature)

• 250 bar / 620°C / 620°C (supercritical, 635°C receiver outlet temperature)

The combination of the receiver and power blocks leads to twelve concepts that are analyzed in detail. The solar fields of the resulting twelve concept variations are analyzed with the heliostat layout program HFLCAL using solar multiples (SM) in a range between 1.5 and 3.5 (intervals with 0.5 SM steps). The resulting 60 heliostat fields are optimized with a genetic algorithm towards minimized levelized heat costs (LHC) and then applied for representing annual calculations on an hourly basis. With appropriate receiver models the receiver efficiency-load characteristics of the different receiver types with differing irradiation profiles were determined. Then thermal storage capacities representing a full load heat source for the steam generator (SG) between 1 h and 20 h (intervals with 1 h steps) were assumed to optimize the storage capacity towards minimized LEC. For the annual assessment power block efficiency-load profiles provided from a steam turbine manufacturer with associated cost estimations were used. As the provided power block input is confidential, in this paper the relative economic comparison of the concept variants is made.

2. Overall plant specifications

The reference concept for the comparison principally refers to today's state of the art commercial STTP Gemasolar (20 MWel) assuming an increased power level to 125MWel. The site of the comparison is chosen to be Barstow (34.85 N / 116.8 E / 600 m above sea level / DNI at DP = 901 W/m2) while the design point (DP) is chosen to be March 21, 12:00h. The main specifications of the reference concept are given in Table 1. A sketch of the reference concept is shown in Fig. 1.

Table 1. Main specification of the reference STTP concept for the solar field

Solar Field

Type 360° (surrounding field)

Reflective area of one heliostat 121 m2

Eff. reflectivity 89.34 %

Beam error 3.664 mrad

Used solar multiples 1.5; 2; 2.5; 3; 3.5

Receiver

Type 360° tubular receiver with external irradiation

Heat Transfer Medium (HTM) Solar Salt / 60 % NaNO3 - 40 % KNO3

Inlet/ Outlet Temperature 290°C/ 565°C

Tube Coating Absorptivity/ Emissivity 93 %/ 83 %

Heat Transport System

Heat Tracing 1.5 % of the generated gross electricity [7]

Thermal Storage System

Type Two Tank

Heat Transfer Medium (HTM) Solar Salt / 60 % NaNOs - 40 % KNO3

Power Block

Type Subcritical Steam Turbine Process

Life Steam Pressure / Temperature 162 bar / 550°C

Fig. 1. Sketch of the reference concept.

3. Receiver concepts and model references

3.1. External Tubular Receiver (ETR)

The reference receiver concepts is the state of the art modular 360° cylindrical ETR, as described in [3]. For the ETR the model described in [3] was slightly modified (number of panels and selective tube coating). The model is then used to gain the efficiency-load characteristics of the receiver, considering an appropriate irradiation profile provided by the heliostat field layout tool after optimization.

3.2. Internal Tubular Receiver (ITR)

The internal tubular receiver concept uses the assumption, that the absorber tubes, which are similarly arranged in serpentine flow-through panels compared to the ETR, whereas the panels consists of parallel tubes. The difference to the ETR is that these panels are located inside of a cylindrical cavity and at its internal lateral surface. The irradiation of the absorber tubes falls into the cylinder through a circular downwards facing open aperture, while the opposite side of the cavity is closed to reduce free convection losses. The receiver model calculates convective and radiation losses with average temperatures. The calculations of radiation losses of the ITR require the knowledge of the net heat transfer between the surrounding surfaces. In this study the Enclosure Method of the VDI Heat Atlas [8] is extended to consider solar radiation on involved surfaces. Free convective losses of the cavity are taken into account with the correlation by Paitoonsurikarn [9].

3.3. Internal Direct Absorption Receiver (IDAR)

Recently a new receiver concept with a directly irradiated liquid molten salt film as the coolant of inclined absorber walls was introduced [5]. The developed IDAR-CFD-Model was used to analyze the open parameters concerning the feasibility and functionality of the concept and considers the optics of the open free film surface and of the inclined opaque absorber wall, as well as the convective heat transfer between the absorber surface and the liquid film. The model allows to carry out the required detailed calculations at full size receiver geometries. The detailed description of the used IDAR model is available in [5] and is used for this study.

Fig. 2. Sketch of the receiver concepts.

4. Methodology of the concept assessment and tools

The used tools for the entire concept assessment are HFLCAL [10] for the heliostat field layout optimization and SPRAY [11] for the ray-tracing from the heliostat field to the receiver. The optimization of the heliostat field leads to the concentrator configuration with optimal cost, which comprises the estimated optimal number of heliostats, an estimated optimal height of the solar tower and the estimated optimal area of the receiver aperture. ANSYS-CFX is used for the fluid mechanical and thermodynamical modeling and optimization of the IDAR. The receiver models of the ETR and the ITR are less detailed and the analytical and numerical interrelations are implemented into ExcelSheets with additional VBA-Macros. FEMRAY, an internal DLR tool, is used for the adaptation of the irradiation data to the CFD-Mesh of the IDAR. The receiver models use the irradiation flux distribution on the absorber surfaces of the particular receiver concept to calculate the receiver efficiency matrix for a given optimal concentrator system and for given geometrical and thermal boundary conditions. The receiver efficiency matrix is used for the annual performance assessment to interpolate the receiver efficiency for each incident radiation and ambient temperature of each daytime of the considered year. For the annual performance of the considered solar power plant variations the ECOSTAR methodology was modified to assess the performance of solar towers feeding supercritical steam cycles. It calculates the annual electricity production hour by hour, taking into account the available solar radiation, the load curve, the part load performance of all components and the ambient temperature. The parasitic energy requirements and the operation of a thermal energy storage are also considered. The entire model for LEC comparison uses a solar only operating strategy (no fossil co-firing) and common assumptions for the site, meteorological data and the load curve. Furthermore a simple cost model according to the model suggested from the International Energy Agency [12] is used.

The approach is to analyze the cost reduction potential of different STP concepts for supercritical steam cycles compared to a reference system. The term of merit is the LEC but also annual efficiencies and yields are compared to identify future research tasks that have the highest potentials to lower the LEC of STTP.

5. Results

In Fig. 3 the relative LEC comparison of the reference concept can be seen. The ten curves represent the results of the annual calculations for the mentioned five different SM. Each curve consists of five data points, whereas the data point in the middle represents the optimal storage capacity referring to the particular SM and site. For a general comparison also the results of Almeria as a potential European site are given. The reference LEC (100 %) refers to a state of the art STTP with a SM of 2.5 and an optimal storage capacity of 11h at the site of Barstow.

4 6 8 10 12 14 16 opt. storage capacity[h]

Barstow Almeria

-ETR-T565-SM1.5-subc > ETR-T565-SM1.5-subc

□ ETR-T565-SM2.0-subc □ ETR-T565-SM2.0-subc

"ETR-T565-SM2.5-subc * ETR-T565-SM2.5-subc

ETR-T565-SM3.0-subc O ETR-T565-SM3.0-subc

-ETR-T565-SM3.5-subc I ETR-T565-SM3.5-subc

Fig. 3. Relative LEC comparison of the considered state of the art STTP over the considered storage capacities

for different solar multiples and sites.

To gain a more deep insight related to the heat balance of the components, Fig. 4 depicts the annual energy transferred from one component to the other. It is obvious, that the annual yields grow with increasing SM. The annual efficiencies of the considered reference components are also shown in Fig. 4, whereas the annual total net efficiency of the plant is decreasing with increasing SM.

Fig. 4. Annual energy yields of the reference with varied solar multiples (left) and annual component efficiencies of the reference (right).

For this, with each of the twelve concept variations the variation of the SM and the storage capacity was carried out. The curves with the lowest LEC referring to an optimal SM are shown in Fig. 5. This way, the optimal reference system (1) can be compared to the eleven other optimized concept options comprising the innovative receiver and power block components. The calculations were carried out with the same specific cost estimations of all components, expect of the power block. For the power block a turbine manufacturer provided different costs depending on the temperature and pressure level. The interpretation of the results depending on the cost assumptions will take place in the section Sensitivity Study.

With the same specific cost assumptions of heliostats, receiver, storage and the same exponential cost correlation for the tower costs, the results show, that all concept variants differ from each other less than 5 %. No LEC reduction potential can be expected with supercritical steam cycles or temperature increase. This is mainly caused by the made power block cost assumptions, whereas supercritical steam cycles are predicted to be significantly more expensive than subcritical steam cycles, especially, when the life steam temperature is raised over 600°C. The specific cost increase even of the low temperature supercritical power block option is too high to be able to lower the LEC under the one of the reference. The underlying cost assumptions are listed in Table 2.

Table 2. Specific cost assumptions or correlations

Component Costs

Heliostats 140 €/m2

Tower 250000 € + 14.77 € • HT239

Receiver 125 €/m2

Thermal Storage 27.5 €/kWhth

Power Block / 162 bar / 550°C (subcritical) 100 %

Power Block / 162 bar / 620°C (subcritical) 118 %

Power Block / 250 bar / 550°C (supercritical) 112 %

Power Block / 250 bar / 620°C (supercritical) 129 %

Indirect costs (surcharges for construction, engineering and risks) 30 %

The results in Fig 5 show furthermore, that a very low potential for LEC reduction (0.5 %) exists for the two cavity receiver options (concept 2 & 3), if they fed a steam cycle with reference steam parameters. At the made cost assumptions none of the here introduced innovative concepts show potentials for LEC reduction compared to the reference.

106% 105% 104%

< 103%

external tubes

^ETR-T565-SM2.5-subc (1) ] ETR-T565-SM2.5-supc (4) ) ETR-T635-SM2.5-subc (7) — ETR-T635-SM2.5-supc (10)

8 10 12 opt. storage capacity[h]

internal tubes

k- • ITR-T565-SM2.5-subc (2) H ITR-T565-SM2.5-supc (5) » ITR-T635-SM2.5-subc (8) K •ITR-T635-SM2.5-supc (11)

internal film

• IDAR-T565-SM2.5-subc (3) IDAR-T565-SM2.5-supc (6) IDAR-T635-SM2.5-subc (9) IDAR-T635-SM2.5-supc (12)

Fig. 5. Relative LEC comparison of the considered STTP over the considered storage capacities for different solar multiples and sites.

To be able to observe the concepts more in detail on an annual basis, Fig. 6 contains the annual efficiency data of the analyzed concept variants. Note that in Fig. 6 the concept numeration accords to the numbered legend in Fig. 5. It is obvious, that the cavity receiver concepts (2, 3, 5, 6, 8, 9, 11 and 12) show an increase in the receiver thermal efficiency, which raises the combined efficiency of the solar components (Field, Receiver and Storage) of the cavity concepts by ~3 %-points. With this the cavity receiver concepts with downwards facing aperture reach total net efficiencies that are between ~6 % and 7 % higher than the reference receiver concepts.

50% -40% -30% -20% -10% -0%

1 2 3 4 5 6 7 8 9 10 11 12

□ n Field 0.563 0.565 0.567 0.560 0.563 0.563 0.552 0.564 0.555 0.554 0.563 0.562

O n Receiver 0.836 0.893 0.891 0.832 0.891 0.891 0.838 0.879 0.879 0.837 0.879 0.877

□ n Solar 0.470 0.504 0.505 0.466 0.501 0.501 0.462 0.496 0.488 0.464 0.495 0.493

□ n PB net 0.410 0.408 0.408 0.412 0.410 0.410 0.424 0.422 0.422 0.429 0.427 0.426

□ n total net 0.191 0.204 0.205 0.191 0.204 0.204 0.195 0.208 0.205 0.198 0.210 0.209

Fig. 6. Annual efficiencies of selected components.

In Fig. 7 the relative differences between the field size, the annual yields, the total investment costs, the total net efficiency and the LEC are compared. At first sight one is wondering, that a concept with higher total net efficiency by 6.7 % (e.g. comparison between concept 1 and 2) and lower total investment costs by 2.1 % leads to a LEC reduction not higher than 0.5 %. But as for the LEC calculation also the annual yield and the investment costs are significant parameters, this effect needs to be explained more in detail.

As optimized STTP power plants are compared to each other and the annual yield of the STTP is mainly defined by the power block (125 MWel) and a solar multiple (1.5-3.5), which by definition lead to the receivers thermal power level at the DP. Because of this reason the resulting cost optimal field size (number of heliostats) of the concepts variants is differing. It seems likely, that a concept with a higher total net efficiency requires less heliostats to fulfill the same set demand at the DP. This again leads to different annual yields of the concepts that are significant for the LEC calculation. Comparing again the concepts 1 and 2 in Fig. 7 focusing on the difference in field size and annual yield, it is apparent that the field size of the ITR concept is by 7.7 % lower. Due to the lower number of heliostats the investment costs decrease by 2.1 %, but contrariwise also the annual yield decreases by 1.6 %. In sum only an LEC reduction of 0.5 % remains between concept 1 and 2.

Fig. 7. Significant relative differences between the assessed concept variations.

Remarkable is the fact, that all concept variation expect of concept 4 (ETR with low temperature level and supercritical power block) disclose a significant increase in total net efficiency, but also, that the optimal field sizes shrink proportionally. As the benefit of increased total net efficiency provided in our study is in agreement with the previous studies, the only reason, why our LEC results are different, lies behind differing cost assumptions.

Fig. 8. Relative component specific investment costs

The relative difference of the component specific investment costs are depicted in Fig. 8. Here it will be obvious, that of course if the investment costs, mainly because of the power block investment assumptions, almost remain the equal compared to the reference, but the annual yield even decreases, no LEC reduction potential can result, even when significantly higher total net efficiencies can be reached with the innovations. The different cost assumptions explain also, why Kelly predicts no LEC reduction potential for ETR with supercritical steam cycles and temperature increase to ~600°C, while Kolb and Singer predict for the same an LEC reduction potential of ~5 %.

To relativize the gained results, sensitivity studies were carried out to be able to observe the dependency of the cost assumptions on the LEC. This was done with the costs of all components, applying lower (factor 0.85) and higher (factor 1.15) component specific investment costs. One example is shown in Fig. 9 that refers to the costs of the heliostats. Connected to the variation of the heliostat costs, two main statements can be done. On one hand the heliostat costs of ITR or IDAR based systems have a slightly lower sensitivity on LEC than the ETR based concepts. On the other hand, comparing the gradients of the heliostat cost and power block cost sensitivity analysis, the heliostat costs have the highest sensitivity on LEC directly followed by the power block cost estimates.

126 133 140 147 154

spec. heliostat field costs [EUR/m2]

external tubes

- RoRe-T565-SM2.5-subc

- RoRe-T565-SM2.5-supc

- RoRe-T635-SM2.5-subc

- RoRe-T635-SM2.5-supc

internal tubes

-♦- CanRe-T565-SM2.5-subc

— A- CanRe-T565-SM2.5-supc

— CanRe-T635-SM2.5-subc

— •- CanRe-T635-SM2.5-supc

internal film

• IDAR-T565-SM2.5-subc

• IDAR-T565-SM2.5-supc ■ IDAR-T635-SM2.5-subc

• IDAR-T635-SM2.5-supc

Fig. 9. Sensitivity analyses of the specific heliostat costs.

Related to the tower cost assumptions the statement can be done, that an ETR based STTP will reach a cost reduction of ~0.5 %, if the tower costs can be reduced by 15 %, while the same relative tower cost reduction leads in the case of an ITR or IDAR based STTP to an LEC reduction of ~1 %.

Varying the cost assumptions of the receiver and assuming 15 % lower values, the LEC reduction is ~3 %. If the statement of Wu [13] is valid, that due to no existent tubes a direct absorption receiver with liquid film cooling could be ~30 % lower than the one an ETR, the cost reduction potential of an IDAR (565°C, subcritical PB) could ensure a LEC reduction potential of ~6 %, which is in close agreement with Singer [5].

The sensitivity curves of the thermal storage system and the O&M cost assumptions run almost parallel to the sensitivity curves of the receiver.

Related to the actual power block cost assumptions, the statement has to be made that neither the temperature increase, nor the pressure increase and also both cannot lead to a cost reduction, as these types of power blocks will be too expensive for the next generation STTP. Additional cost assumptions from the turbine manufacturer state, that in case of a temperature increase of the Rankine process, nickel based alloys could be necessary mainly because of corrosion reasons. If this is the case, the power blocks with increased temperatures had instead of 18 % or 29 % higher costs than the reference power block a significantly higher offset of 67 % and 78 %. With this the chances of supercritical STTP with increased temperatures seem to shrink.

6. Conclusions

After the discussion of the results of this study the following conclusions can be made:

• With the made technical and economic assumptions, which comprise the same component specific cost estimates and correlations expect of the power block variants, neither the temperature increase, nor the pressure increase of the STTP's Rankine cycle would lead to a reduction of the LEC compared to the state of the art.

• The only and not significant cost reduction potential of ~0.5 % can be observed, if ITR or IDAR type receivers are used and the reference Rankine cycle is not changed.

• If the cost estimations of SNL related to DAR with liquid film cooling apply, an IDAR could have specific receiver costs, which are ~30 % lower than the reference ETR or the ITR. If this is the case, a LEC reduction of up to 6 % could be reached by changing the receiver from ETR to IDAR, without a temperature increase or the increase of the steam parameters.

• The application of power blocks with increased steam parameters and thus increased thermal efficiencies leads to increased total net efficiencies of the entire STTP. However, with the used power block cost assumptions this effect doesn't result in reduced LEC.

References

[1] Kolb GJ. An Evaluation of Possible Next-Generation High-Temperature Molten-Salt Power Towers. Sandia National Laboratories, SAND2011-9320, Alberqueque, USA; 2011.

[2] Kelly BD. Advanced Thermal Storage for Central Receivers with Supercritical Coolants. Abengoa Solar Inc. Technical Report, DOE/GO18149, USA; 2010.

[3] Singer Cs, Buck R, Pitz-Paal R, Muller-Steinhagen H. Assessment of Solar Power Tower Driven Ultrasupercritical Steam Cycles Applying Tubular Central Receivers With Varied Heat Transfer Media. Journal of Solar Energy Engineering, Vol. 132, 2010, 041010-12.

[4] Singer Cs, Buck R, Pitz-Paal R, Muller-Steinhagen H. Economic Potential of Innovative Receiver Concepts with Different Solar Field Configurations for Supercritical Steam Cycles. 5th International Conference on Energy Sustainability, 07.-10. 2011, Washington DC.

[5] Singer Cs, Buck R, Pitz-Paal R, Muller-Steinhagen H. Economic Chances and Technical Risks of the Internal Direct Absorption Receiver (IDAR). 6th International Conference on Energy Sustainability, 23.-26. Jul. 2012, San Diego, California.

[6] Pitz-Paal R, Dersch J, Milow B, Tellez F, Ferriere A, Langnickel U, Steinfeld A, Karni J, Zarza E, and Popel O. Development Steps for Parabolic Trough Solar Power Technologies With Maximum Impact on Cost Reduction. Journal of Solar Energy Engineering, 129(4), pp. 371-377; 2007.

[7] Pacheco JE, Bradshaw RW, Dawson DB. Final Test and Evaluation Results from the Solar Two Project. Sandia National Laboratories, SAND2002-0120, Albuquerque, USA, 2002.

[8] VDI Heat Atlas. 10th Edition; 2006.

[9] Paitoonsurikarn S, Lovegrove K. A new correlation for predicting the free convection loss from solar dish concentrating receivers. 44th ANZSES Conference, Australia; 2006.

[10] Schwarzbozl P, Pitz-Paal R, Schmitz M. Visual HFLCAL - A Software Tool for Layout and Optimisation of Heliostat Fields. 15th International SolarPACES Symposium, 15.-18. Sept., Berlin; 2009.

[11] Buck R. Solar Power Raytracing Tool SPRAY. German Aerospace Center (DLR), User Manual -Version 2.6, Stuttgart; 2010.

[12] International Energy Agency (IEA). Guidelines for the Economic Analysis of Renewable Energy Technology Applications. Publication; 1991

[13] Wu SF, Narayanan TV. Commercial direct absorption receiver design studies. Foster Wheeler Solar Development Corp., SAND-88-7038, Livingston, USA; 1988.