Analysis of a Concentrating Solar Power Tower Operating with a Closed Joule Brayton Cycle and Thermal Storage

The future of our planet depends heavily on protection of the environment and the expansion of the use of renewable energy [1]. In these years the interest in concentrating solar power tower (CSP) is strongly growing, and it is extending also to Italian areas. Actually, in the world, CSP plants produce around 483.6 MW, and 457 MW are at commercial stage, while other 430 MW are in developing [2]. In previous works [3,4] the use of air as working fluid in solar power plants has already been treated.

In this paper, a solar thermal power plant, with a Joule-Brayton closed cycle, employing air like fluid work is presented. Figure 1 shows the system, consisting in the CSP plant, a gas turbine as power generation engine, a thermal storage system (TES), cooling subsystem and auxiliary devices.

The CSP plant is composed by the concentrating solar tower and the heliostats field, with a North field configuration, which, through a two axes rotation system, allows conveying the solar radiation at the top of the tower. A cavity volumetric closed receiver, necessary for this plant configuration, has been placed [5].

The power plant consists of a gas turbine of 5 MW peak power, which uses only air. The TES is made up of an air-salt heat exchanger, two storage tanks ("hot" and "cold") operating at temperatures between 790°C and 450°C and a circulation molten salt pump. The thermal storage medium is a mixture of molten salts KCl-MgCl₂, whose characteristics are shown in Table 4. Lastly, in the system there are a heat exchanger, which cools down the air temperature at initial conditions of the cycle, an auxiliary compressor and a vent valve, which regulates the pressure, and a cooling tower.

The climate data referring to Seville, for the annual simulations, have been used and they have been extrapolated from TMY3 [6]. The aim of this work is to establish the optimal value of the solar multiple (SM) with the storage capacity (hours at full load) in order to minimize the value of the LCOE.

Annual hourly simulations have been conducted using Thermoflex [7] and varying the hours of thermal storage as well as the SM. The sizing of the heliostat field and the optical efficiency matrix have been obtained by software WinDelsol [8]

figure_1.png

Figure 1. Plant scheme: 1. Auxiliary compressor; 2.GT compressor, 3. Solar tower; 4.GT turbine; 5. Cold tank; 6. Hot tank; 7. Storage heat exchanger; 8. Salt pump; 9. Bleed valve; 10. Pre-cooler; 11. Cooling tower.

After the design of the heliostats field, by WinDelsol software, setting the radiative and convective losses, the efficiency matrix, function of azimuth and solar altitude, have been extrapolated. For each SM configuration, and using this matrix, an hourly inlet thermal power in the solar field has been calculated. For example, in Figure 3 the matrix of efficiency for SM = 1 is reported.

figure_3.png

Figure 3. Efficiency Matrix of solar field for SM=1; X axis solar altitude, Y axis Azimuth

The availability of hourly power data allows the evaluation of the annual inlet thermal energy, for each configuration. The storage tank volumes, for a ranging time from 2 to 8 hours, using the salt properties listed in Table 4, were calculated. The resulting volumes, and the salt masses, are reported in Table 6.

Table 6. Storage volume and salt mass for different hours of storage.

Due to the variability of the mass flow of the working fluid, that follows the DNI variation, the salt mass flow rates from the tanks change. The values required, to accommodate the different load conditions, have been evaluated.

3.1 Simulation

The annually energy production, for SM=1 configuration, is of 12.50 GWh. Changing the SM, for values ranging from 1.2 to 2.4, the net electric energy produced and plant costs for the different sizes have been evaluated.

Figure 4 shows the energy production for all configurations (SM and hours of storage); the best configuration can be selected using the minimum value of LCOE.

figure_4.png

Figure 4. Energy production [GWh], x axis SM, y axis TES hours, z axis net electric energy

The result appears trivial since the energy production increases with both the hours of storage and the SM. In particular, it increased from 12.50 GWh for SM = 1 to a maximum value of 26.33 GWh for SM = 2.4.

Therefore, the choice of the best configuration is not about only the energy production. It is important to analyze other parameters, like the lost energy for defocusing; this energy cannot produce useful effects, neither for the PB nor for TES, due to their limited size.

Figure 5 shows the lost and produced energy during a year time.

figure_5.jpg

Figure 5. Lost energy for defocusing and energy produced for different plant configuration.

As shown in Figure 5, it is evident that the increasing of SM causes higher lost energy, so an asymptotic growing trend of the electrical energy production curves is confirmed.

Another parameter, for the choosing of the optimal system configuration is the thermodynamic efficiency, expressed in equation (6):

$\eta_{s y s t e m}=\frac{E_{n e t}}{E_{S F}}$(6)

Table 7. Thermodynamic efficiency of the system for different system configurations

The data analysis obtained by the equation (6), shows how the best configuration is SM = 1.2, for all the storage capacities. However, the most important parameter about evaluation of the best convenient configuration plant is LCOE, calculated in this work as shown in equation (7):

$LCOE=\frac{\alpha \times {{C}_{inv}}+{{C}_{ma\operatorname{int}}}+{{C}_{fuel}}}{{{E}_{net}}}$ (7)

For the LCOE analysis, the cost values shown in table 8, 9 have been utilized.

Table 8. Investment cost [M USD] for different plant configuration; x axis hours storage, y axis SM

Table 9. Unit cost using in the economic analysis [14][15][16] [17][18]

The results of the LCOE analysis are shown in Figure 6.

figure_6.jpg

Figure 6. LCOE analysis for different SM and hour storage

In figure 6 it is possible to note that the minimum value is obtained for SM = 2.0 and 8 hours of storage capacity. In this configuration, the LCOE value is 15.1 ¢$ / kWh.

A technical and an economic analysis for a CSP plant with GT and TES have been conducted. The pressure control system allows to achieve consistent performance, in different operating conditions, without fuel consumption. The use of the mixture of molten salts KCl-MgCl₂ is very innovative for the high melting temperatures and for low specific costs, which allows using effectively with the power system proposed. The use of the storage has then allowed to increase the capacity factor of the plant, as well as the operation hours, in the same way as the current solar power systems. The trend of the economic analysis, founded on LCOE method, confirms how the unit cost of energy produced by the plant configuration here presented is similar to the hybrid CSP plants in commercially stage [15]. In conclusion, the study of the utilization of the solar field, with tower, receiver and storage system, provides a very attractive value of LCOE of 15.1 ¢$/kWh, in a solar power plant characterized by SM= 2.0 with a storage having 8 hours capacity. This solution has a value of CF of 55.95%, which is not far from of CF values achieved by typical commercial plants, and higher than the photovoltaic system ones. Considering that this LCOE result was obtained without accounting any incentives, hopefully this technology soon will achieve higher levels of competitiveness in the world energy market.