Technical and Financial Viability of a 1 MW CSP Power Plant with Organic Rankine Module: Case Study for a Northeastern Brazilian City

In this study, a 1 MWe parabolic trough concentrating solar power plant using an Organic Rankine Cycle to convert thermal power into electricity was simulated. The solar data used is from Fortaleza, the capital city of Cear´a, a state in the northeast region of Brazil. The simulation was year-round and used hourly acquired data. Several different configurations differing in number of collectors and size of thermal energy storage were considered and, by the end of the study, according to the technical and financial results, one best configuration was defined. With a net annual energy generation of 4815.50 MWh, a 12.08% total efficiency and a levelized cost of electricity (LCOE) of 186.98 US $ /MWh, the best defined configuration was the one with 100 collector assemblies and 48 MWh storage system. Besides that, the capital and OM costs of the various components were compared to define the most expensive to the power plant, being the collectors acquisition the biggest cost, which represents 74% and 58% of capital and OM costs respectively.


Introduction
In today's context, renewable energy sources are playing a bigger role in electricity generation than years ago.Between 2016 and 2017, the world has seen an increase of 178 GW of renewable generation installed capacity, a raise of 8.82%.In a somewhat different way, concentrating solar power is still growing in a short pace, 0.1 GW from 2016 to 2017, reaching 4.9 GW, which represents only 0.22% of the total renewable capacity in that year [1].Even though the investments in CSP are still low when comparing to wind and photovoltaics, it is still one of the most promising technologies to help in the transition from the traditional thermal generation to the renewables, mostly due to the possibility of decoupling the energy production from the solar source with the use of an energy storage system [2].Concentrating solar power consists of harvesting the energy from the Sun's radiation in the form of heat for electrical power production, by concentrating the direct solar radiation onto a receiver and using a heat engine for the power conversion.In order to do that, a set of different thermal cycles may be implemented, such as: Rankine or Stirling [3].Typically, for parabolic trough collectors, the Rankine cycle is implemented [3] and it should be used as a baseline comparison for the present paper.The main features that a site must have to be suitable for such a power plant are abundant direct sunlight (DNI, direct normal irradiation) during most of the year and high average temperatures to reduce heat losses to the ambient.Solar power is inherently intermittent and, due to that, CSP can be implemented together with a thermal energy storage (TES) solution, a system responsible for storing the exceeding thermal energy production in the peak hours and using it during cloudy or night hours.The TES allows the CSP power plant to have a greater efficiency and availability [4].Although this technology has been the topic of studies for several years [5], it still is a field presenting continuous developments, as investigated by [6] who analyses the present status and the development tendency of CSP, and being the focus of many recent researches which address many issues related to it.In [7] a fully renewable system comprising wind, CSP and biomass power plants, and pumped-storage facilities was analysed.The authors presented CSP as crucial for the operation of such a system, due to the fact that plants with TES are dispatchable and can adjust their production to the demand, providing more flexibility to the system.[8] investigates the feasibility of CSP in Sub-Saharan Africa, and concludes, among other things, that the financing conditions are the most important factor to determine the CSP feasibility for that region.In [9], it was analysed the economic justification of CSP when taking into account not only the selling price of energy but also the flexibility provided to the system and the economical gain due to it in two provincial power systems with high renewable penetration, in northwestern China.Conclusions could be drawn, like that CSP investment is more cost-efficient in systems with deficient operational flexibility, and that substituting variable renewable energy with CSP brings considerable reductions in thermal generation fuel costs, ramping costs, and start-stop costs, in which the fuel cost reductions are the dominated contributor.When it comes to Brazil, the CSP technology still must carve its way inside the country's energy matrix, since in the past decades all the investments in renewable capacity went to hydro, wind, PV or biomass generation [10].Being near the Equator line, Brazil's northeastern region presents a good potential for CSP projects, with an average DNI reaching around 5500 W h/m 2 .day and average annual temperatures superior to 24°C [11].Furthermore, the region has seen a great rise of wind and solar power generation [10], which brings instability to the system and makes CSP projects more attractive for the system operator in order to compensate the typical PV and wind variations with a dispatchable and stable renewable capacity [9].This rise can be seen in figure 1 where the power generation by source of the region is shown for the period of 2010 until 2018.Taking this local favourable conditions into account, in this paper a 1 MWe CSP power plant with an Organic Rankine Cycle (ORC) power block is proposed and the technical and economic results of a year-round operation are estimated, analysed and compared with values of Levelized Cost of Electricity (LCOE) presented as trend in [1] and with the results obtained in [12] for a hybrid CSP and biomass power plant also in the northeastern region of Brazil.In order to do so, several configurations were considered, differing one from another in the solar field size and in the implementation or not of a TES and the amount of its storage capacity.The LCOE was calculated for a 30 years operating period and the optimum configuration is chosen based on the lowest result.The present work utilizes local sunlight data of a Brazilian northeastern city for the calculations, thus reaching results that should approximate to the reality of this region that doesn't have any working plant of the kind in its territory, despite the good potential it presents.This has been investigated also by [12], where a hybrid CSP and biomass power plant is simulated for a region of the state of Bahia in the northeastern region of Brazil.Furthermore, this study raised the Organic Rankine Cycle (ORC) technology as a suitable option to allow a relatively efficient operation in lower temperatures and for a small capacity, as was done in [13], where a model to analyze the impact of different design parameters on the LCOE of a 1MWe plant is proposed.The ORC application for such a power plant was also analysed by [14], although for smaller systems (up to 50kW).The technology chosen for the solar field was the parabolic trough collectors (PTCs), once it is the most commercially available and tested among the thermal collector options, presenting higher efficiency than the solar tower and Fresnel technologies [3,15].The novelty of the present work relies on many of its aspects.At first, the chosen location, a Brazilian northeastern city, has not been studied for this kind of undertaking, even though it presents two important aspects to attract investments in the field: the region has favourable weather conditions for the presented technology and it has a large population, causing it to be a high demand location for the transmission system.Together with the mentioned aspects, the brazilian integrated system already faces a challenge that should only increase in the years to come: the instability brought to the grid by renewable capacity from wind and solar, causing this kind of investment to be even more attractive for the operator.Given all that, sunlight data measured in the city of Fortaleza was applied in the calculations here performed.Another novel aspect of this study is the utilization of a real commercially available ORC module for the power conversion with the goal of allowing the implementation of a small capacity power plant (1M W e) but with high flexibility and low maintenance costs.This can be a good alternative for a first investment of the kind in the region, since the discussion is centered in a region that has no precedent in the CSP business.The third novelty presented here is the sizing of the solar field in function of the chosen ORC module in order to obtain the best configuration for the plant regarding solar field size and storage capacity with the goal of obtaining the lowest LCOE possible given the chosen constraints.Through the economical analysis, it was possible to state the components that influence the most the obtained energy price and to understand how the storage capacity and the solar multiple affect the LCOE of each configuration.This paper should foster discussions and further studies for the region regarding this kind of technology in order to develop a scenario more prone to receive investments from companies and institutions interested in further developing the renewable energy sector in Brazil and, more specifically, its northeastern region.Currently this region doesn't have a wide enough background to rely on or to draw the interest of such players.In section 2 of this work, the system modeling is described for each of its components with all the physical and empirical equations used and the assumptions made.Section 3 as used to present the parameters considered for the economic analysis of the power plant and how the LCOE was calculated.An overview on the simulation is given using an hourly operation graph for one day and a sankey diagram for one specific hour of the day.In section 5, the results are presented and discussed and the final conclusions are drawn in section 6.

System Modeling
The model was applied utilizing Sun irradiation data acquired at the airport of Fortaleza city (3°46' 33" S 38°31' 56" W), the capital of Ceará, a northeastern state of Brazil.The data provides hourly values for Direct Normal Irradiation (DNI) throughout a year and is available as weather data from the software System Advisor Model by the National Renewable Energy Laboratory (NREL) [16], which acquires those values through ground measurements, satellite data or both.The solar field model was based on [17] and [5], where all the equations needed to calculate the necessary angles and the thermodynamic performance of the solar field were taken from.The collector considered was the LUZ LS-2, the same model implemented in some of the SEGS power plants in the Mojave Desert in California [5].The option for this collector model enables better comparison possibilities between the results obtained and the vast amount of data and studies based on the SEGS power plants.In the next subsections the mathematical description of the different subsystems is described.The subsections 2.1, 2.2 and 2.3 present the equations and assumptions used for the power block, the solar field and the thermal storage system respectively and the subsection 2.4 describes the different configurations considered in this paper.

Power Block
The main constraint of the system is the ORC module, since it defines the amount of power being generated according to the thermal energy supplied at the evaporator and the temperature at which this energy is supplied thus regulating the energy transferred to and from the TES.The module considered in this paper is the same for all the 10 configurations and generates a net electric energy of 964 kW with 4043 kW of thermal energy from the HTF, reaching a 23.8%net efficiency at base load [18].For operation under base load conditions, a linear drop of 1% in efficiency for every 5% in load was considered, according to the efficiency curve presented in [18].Table 1 summarizes the main technical data of the ORC module.As can be found in the literature [17,19], the percentage of the gross power production that is consumed by the cooling water pumping system lies around 2.26%.Given that, a percentage of 2.5% was used as a reasonable assumption for the present work.The variation of ambient conditions was not considered to cause greater impact in the operation of the cooling system, being thus neglected.

The Solar Collectors Field
Before any mathematical modelling, it's important to determine the design point for the given solar power plant.Usually, the design point irradiation is taken at equinox at 10am [20].For 2019, the equinox was on 23rd of September so the design point for the present paper will be considered that date at 10am.Starting with the absorbed radiation equation, the solar field model determines how much energy the collectors are harvesting from the Sun.Equation 1 calculates the absorbed energy as a function of the Direct Normal Insolation (DNI) and many other factors that tend to lower the fraction that is really absorbed: Each of the factors presented in the equation is discussed separately in the sections 2.2.1 until 2.2.4 and the necessary equations for estimating them are presented.All these equations were taken from the literature [17].Heat losses are modelled in subsections 2.2.5 and 2.2.6, the energy gain equation for the HTF is presented in 2.2.7 and sections 2.2.8 and 2.2.9 address the power consumption of the HTF circulating pumps and how the HTF flow rate control is carried, respectively.

Incidence Angle Modifier (IAM)
The incidence angle not only influences the absorbed radiation through the cosine present in the equation but also through other losses that must be considered.They occur due to additional reflections and absorptions by the mirrors as a result of the increment in the incidence angle.The model assumed for these losses results from empirical data of the LS-2 collectors and is formulated according to the equations 2 and 3. (2)

Shading
The shading due to parallel adjacent collector rows occurs in the early hours of the day and late hours of the afternoon, and it is estimated by equation 4. The effective wideness is the portion of the aperture that receives sunlight, that means, it is not covered by shadow.

Heat Collecting Element (HCE) End Losses
The end of the HCE is responsible for a heat loss, since that, depending on the incidence angle, a portion of the element won't receive radiation due to the collector structure.The equation 5 estimates a value for that factor:

Solar Field and HCE Efficiencies
The last factors to be considered account for mirror imperfections, dirtiness, reflectivity of the HCE glass, among others.They are modeled by equations 6 and 7:

Heat Collector Element (HCE) Heat Losses
For the HCE heat losses, the model developed by [21] and utilized in [17] was also implemented in the present work.Many factors like wind speed, incident radiation, HTF mass flow rate, ambient temperature and bulk fluid temperature can have influence in the heat lost to the surroundings but, to make a simpler model and optimize the computational overhead, the DNI and the fluid temperature were considered as the main influencing factors.With these assumptions and through a linear regression of the calculated heat loss per unit length over a range of bulk fluid temperatures and DNI levels, for a UVAC tube with cermet selective coating, at an ambient temperature of 25 • C with a volumetric flow rate through each collector of 140 gal/min the equation 8 is obtained.Since the temperature varies along the HCE, the equation has to be integrated for the inlet and outlet temperatures.In addition, the result should be divided by the collector aperture in order to represent the heat loss in W/m².Equation 9 is obtained and it is utilized for a single type of HCE and with optimum vacuum conditions.

Piping Heat Losses
The heat losses of the the solar field piping system can be estimated by the empirical equation 10 : where SfPipeHl is expressed in W/m 2 and ∆T is the difference between the average temperature of the solar field and the ambient temperature.Generally, the piping losses are small, reaching values around 10 W/m 2 or less during operation [17].

HTF Energy Gain and Temperature Rise
The energy collected by the solar field is the difference between the heat absorbed and the heat losses presented previously.
The solar field was assumed to operate at steady state conditions and the kinetic and potential energy changes were neglected.The enthalpy variation of the HTF is then calculated by equation 12: The density or the Therminol VP-1 as a function of its temperature is represented by equation 13 [17]: And the temperature and specific enthalpy as a function from one another are represented by equations 14 and 15 [17]:

HTF Pump Power Consumption
The power consumption from the HTF pump was calculated using [17] as reference.In that work, the author had the pump rated power as input and calculated the pressure drop with the equation 16 Here, the inverse path was made, considering the pressure drop as proportional to the number of collectors in the field.Given that, the power consumed was calculated with the pressure drop, the efficiency of the pump and the volumetric flow as inputs.In addition, the efficiency of the pump was considered to be 60% at nominal operation, as in [17], which would be for the 23rd of September at 10am, according to the design point stated at the beginning of the section 2.2, and it was corrected with the flow variation according to equation 17.
2.2.9.Heat Transfer Fluid (HTF)Flow Rate and Outlet Temperature Control During night or cloudy hours, keeping the HTF circulating would result in heat losses without any gain.Given that, it was assumed that the pumps would feed the solar field with HTF only if the absorbed heat surpasses the 75 W/m 2 threshold.Furthermore, the ORC module is specified for a certain temperature for the oil entering the evaporator.In order to meet such a constraint, the outlet temperature of the solar field was specified at 310°C under normal operation and the HTF flow rate keeps changing in order to control that temperature, that is, the volumetric flow rate may increase in higher insolation times.

Thermal Energy Storage (TES)
Since the heat losses for the HTF were accounted for in the solar field, heat losses associated with TES were considered to be too small (0 kW/K) as in [17].All the excess thermal energy collected by the solar field is stored in the hot tank (until it reaches its peak storage capacity) and, as long as it still has energy stored inside of it, the TES will keep the supply of thermal energy to the power block.

Considered Configurations
For this study, all the configurations considered have the same power block specified in Section 2.1.These differed regarding the presence or not of the TES storage capacity.Five different solar field sizes were also considered: 50, 75, 100, 125 and 150 PTCs assemblies.With the chosen design point stated at 2.2 as reference with a DNI of 810 W/m 2 , the Solar Multiple (SM) of these solar field sizes are 1.42, 2.13, 2.85, 3.56, 4.27.The optimum configuration from those was defined based on the economic analysis which is presented and explained further on.

Economic Analysis
In order to define the financial parameters of the power plant, the different components (solar field, TES, power block) are analyzed separately.Typical values found in the literature are considered for both the acquisition and OM costs of the components mentioned.Those values are applied for the calculation of the Levelized Cost Of Electricity (LCOE).

Solar Field Costs
Based on the literature [22], the acquisition cost of the collectors lies usually between 150-200 $/m 2 .In addition to that, an annual maintenance cost of 3.72 $/m 2 is considered [23].

TES Costs
The values for the cost calculations of the TES are based on a molten salt configuration, which can be justified by the solidification temperature of 142 • presented in [24] for a mixture of N aN O 2 /KN O 3 /N aN O 3 .The literature studied proposes an usual capital cost of 26.22 $/kW h for the acquisition of the TES [4] and a OM cost in the range of 10.5 and 12.6 $/M W h [25], being the average of 11.55 $/M W h the value implemented in the present work.

ORC Module Costs
According to information obtained with the manufacturer (Michael Welch 2018, personal communication, 18 June 2018), the acquisition price may be assumed as 1000 $/kW , with a maintenance cost of 2.50 $/M W h.

LCOE Calculation
The LCOE was calculated using the equation 18 from [19] with and insurance cost added to the OM costs of 0.5% of the cost of capital as in: The annuity factor is calculated with the equation 19 from [26]:

Annual Simulation
Once the solar field model was implemented in Engineering Equation Solver (EES) and the TES and Economic Analysis in a spreadsheet, it was possible to predict what the results would be after a year-long operation of the power plant look like.Figure 2 depicts the energy flows beginning with the incident radiation until the electricity produced by the power plant and the thermal energy stored in a sample day.At the first sunlight hour of the day, the radiation is still not enough to justify the startup of the system until around 7am, when the incident radiation reaches values close to 15MW so the HTF starts circulating and the ORC module starts its power production until it reaches base load operation with 1 MW gross output.Between 7 and 8am, the Thermal Power coming from the solar field reaches values superior to the one demanded by the power block at nominal operation, hence providing the overbalance to the TES, which accumulates the thermal power until around 4pm, when it starts providing energy to keep the ORC module in baseload until 11pm.At that point, the electrical output starts decaying until 1am, when the power block is shut down.From the graph, we can see that the power plant functions at base load during the highest demand hours, which are usually between 10am and 8pm, according to [19].The operation day of the graph is an example of a clear sky one, where the TES was needed only to keep production during the darkest hours and not to secure the stable generation during day time.To illustrate the present discussion, Figure 3 shows the Sankey diagram of the same day of the previous plot (Figure 2), but specifically for the operation at 12pm.It displays the energy flows throughout the system for an instant of high solar incidence level, when a great share of the thermal energy harvested is stored at the molten salt tanks.From the total direct incident solar power, 54.8% is harvested by the solar field and 45.2% is lost to the ambient because of a variety of reasons, already highlighted in Section 2.2.For the analysed instant, we have a large overproduction of heat by the collectors which implicates more than 60% of produced heat being stored in the TES while only 39.8% is destined to the power block for power production.The ORC module selected has a 24.7% gross efficiency and 75.3% of the heat supplied in the evaporator is rejected in the condenser.In this section, the behaviour of the power plant in an hourly and in a daily scale was displayed and analysed in order to provide a better understanding of the energy flows through the systems that compose the power plant and the efficiencies associated with the process.In the next sessions, the results of a year round operation will be discussed in terms of technical and economical aspects and conclusions will be drawn regarding the relevant resulting data.

Results and Discussion
Based on what was presented Figure 4 displays the effect of the TES capacity on the LCOE for the different solar field sized configurations.It's possible to observe that the case where a SM of 2.85 has the curve that reaches the lowest points in comparison to the other four curves, whose points in the direction that it is the optimum configuration from the ones considered.After a careful analysis, it is possible to determine the optimum TES capacity for each solar field size and utilize them to summarize the results for each solar multiple and compare the values for the configurations with and without TES.Those resulted in the configurations show in tables 6 and 7, where the main technical and financial results for each configuration are shown.According to the results shown, the most viable configuration is the one with SM of 2.85 and 48 MWh of TES capacity, since it presented the lowest LCOE with the highest efficiency.This result can be understood as, for the specific ORC module chosen for the specific site analysed, the optimum configuration is the one displaying the best financial results from all the considered configurations.With the 1MW ORC module chosen, any extra investment to increase the solar field or TES sizes would, of course, result in an energy production increase but also an even higher cost addition, resulting in a more expensive electricity for that power plant.Figure 5 shows the net energy generated per month and the monthly average.It is possible to see that the first semester months are under the average.This can be explained by the fact that the region typically goes through a high precipitation season during this period, while the second half of the year is marked by dry weather, less clouds and therefore a higher power generation.The relation between the generation in April, the worst month, and September, the best month in energy generated quantity, is 0.46.That proportion shows that the variation throughout the year is not too expressive, when compared to the numbers in order of 0.3 found by other authors [2].This points to the direction of showing the location as a good one for such a type of project, since it would demand less generation compensations from the system operator in the months with less generation.Figure 6 displays the capital cost of the three main systems of the selected power plant, solar field, TES and power block, and also their sum.It is noticeable that the solar field accounts for the biggest share of the whole investment, reaching about 74% of the total capital cost followed by the TES and then the ORC module, with a slightly smaller number.That is a result that is in accordance with the typical numbers displayed by CSP power plants as stated by [4].Figure 7 displays a similar distribution but regarding the O&M costs of those systems.Again, the solar field accounts for the biggest share of around 58% but, what is worth highlighting is the drastically small O&M cost of the ORC module compared to the other two systems, with a share of only 8% from the total annual cost.In the present section it was possible to analyse the technical and economical aspects of each system of the power plant and of the power plant as a whole and compare different possible configurations for the project.Furthermore, the seasonality of the chosen region was analysed in order to determine the variation of the produced energy along the months.

Conclusion
The stand-alone 1MW Organic Rankine Cycle (ORC) CSP power plant was simulated with the local weather data for 5 different solar multiple (SM) values: 1.42, 2.13, 2.85, 3.56, 4.27 and with different thermal energy storage (TES) capacities, starting from nule storage capacity until values as high as 60 MWh.The levelized cost of electricity (LCOE)of each configuration was calculated and it was possible to determine the optimum configuration from the considered ones as 100 collectors and 48 MWh of TES capacity, with a LCOE of 186.98 USD/MWhe.This value is considerably high when compared to bid tariffs of around 60 USD/MWh reported in Chile and Australia [1], the LCOE of the order of 110 USD/MWh achieved [12] for a hybrid CSP/Biomass power plant in the northeastern region of Brazil and the 87.4 USD/MWh obtained for Morocco [13].Nevertheless, as the capital and OM costs of the power plant are analysed, it is possible to highlight the small contribution of the ORC module in those quantities, 12% and 8% respectively and the big majority presented by the solar field, 73.8% and 58% respectively.Given that, the paper points out the possibility of optimization mainly in the solar field, possibly by the implementation of a more recent model of collector.Another important remark is the advantage the TES brings to the feasibility of the project, since the lowest LCOE achieved without such a system was 197.47 USD/MWhe.Regarding the chosen region, it presents itself as a good option for such a project, not only due to the high DNI and average temperature levels, but also because the monthly generated power variations were low, which shows a fair stability of the power plant throughout the year, making it more attractive to the system operator.
HTF enthalpy change between the inlet and the outlet of the solar field [J/kg] ∆T difference between the average temperature of the solar field and the ambient temperature [ • C]] Ẇpump Power consumption of the pump [W ] η f ield field efficiency that accounts for losses due to mirror optics and imperfections [%] η HCE HCE efficiency that accounts for losses due to HCE optics and imperfections [%] η pumpref reference HTF pump efficiency [%] η pump HTF pump efficiency [%] ρ(T 1 ) HTF density evaluated at the solar field inlet temperature [kg/m] Θ incidence angle [deg] a, b coefficients available in the tables 3 to 5 for different annulus vacuum conditions [−] BelShad losses due to shading of ends of HCEs due to bellows [−] C f uel annual fuel costs (equals to 0, since it's a stand-alone solar plant) [$] C invest investment costs [$] C O&M annual operation and maintenance costs [$] ColF rac fraction of collector type in the field [−] DN I direct normal insolation [W/m 2 ] e mo factor for calculating the HTF pump efficiency (assumed 0.4) [−] EndLoss performance factor that accounts for losses from ends of HCEs [−] EnvT rans transmissivity of the glass envelope [−] f focal length of the collectors (5 meters at SEGS VI) [m] f cr annuity factor [%] GeoAcc geometric accuracy of the collector mirrors [−] HCEabs absorptivity of the HCE selective coating [−] HCEdust losses due to shading of HCE by dust on the envelope [−] HCEEmisc miscellaneous factor to adjust for other HCE losses [−] HCEF rac fraction of HCE type in the field [−] IAM incidence angle modifier [−] K constant for calculating the IAM [−] L SCA length of a single solar collector assembly (50 meters at SEGS VI) [m] L spacing distance between collector rows [m] LCOE levelized cost of energy [U S$/M W h] m HT F ref reference HTF mass flow [kg/m 3 ] m HT F HTF mass flow [kg/m 3 ] M irCln mirror cleanliness [−] M irRef mirror reflectivity [−] n number of periods (assumed as 30 years of operation) [years] N SCA number of collector assemblies in the solar field [−] N umCol the number of collector types in the field [−] N umHCE number of HCE types in the field [−] Q abs radiation absorbed by the HCE [W/m 2 ] Q collected energy collected by the solar field [W/m 2 ] r loan rate (assumed 7% according to [12]) [%p.a.] RowShadow performance factor which accounts for the mutual shading of parallel collector rows during early morning and late evening [−] SF Avail fraction of the solar field that is operable and tracking the sun [−] T bulk fluid temperature [ • C] T 0 temperature in the solar field outlet [ • C] T 1 temperature in the solar field inlet [ • C] T rkW stErr twisting and tracking error associated with the collector type [−] V HT F volumetric flow rate of the HTF entering the solar field [m 3 /s] W collector wideness [m] W ef f effective collector wideness [m]

Figure 1 :
Figure 1: Anual power generation by source for the Northeastern region of Brazil between 2010 and 2018[10]

Figure 2 :
Figure 2: Energy flows in the power plant from incident radiation to generated power for a sample day.

Figure 3 :
Figure 3: Sankey diagram for energy flow for a sample day at 12pm.

Figure 4 :
Figure 4: LCOE variation according to installed TES capacity for five different solar multiples.

Figure 5 :
Figure 5: Monthly power generation for the optimum configuration.

Figure 6 :
Figure 6: Capital costs of the power plant for the optimum configuration.

Figure 7 :
Figure 7: O&M costs of the power plant for the optimum configuration.

Table 1 :
[18]cal operational values for energy input and output and general performance of the ORC module TURBODEN[18]12 HRS -1MW without split provided by the manufacturer.

Table 2 :
[17]cal optical parameters and correction values for the solar field and HCE efficiencies[17].

Table 3 :
Coefficients for the Receiver Heat Loss: Vacuum Annulus

Table 4 :
Coefficients for the Receiver Heat Loss: Air Annulus

Table 5 :
Coefficients for the Receiver Heat Loss: Hydrogen Annulus

Table 6 :
Technical results for the different configurations considered after one year of operation

Table 7 :
Financial results for the different considered configurations.