Performance Assessment of a Low-Global-Warming-Potential Solar-Powered Generator–Chiller

Abstract

This article presents a performance assessment of an electrical power and cooling system powered by a parabolic dish collector and using refrigerants with low global warming potential. The study was conducted using energy and mass balances for each component and system. The simulation includes various parameters, such as solar radiation, the focal temperature of the solar collector, the ambient temperature, the power cycle pressure ratio, and the chiller’s evaporation temperature. The results show that the efficiency of the organic Rankine cycle with the refrigerant R1233zd(E) is similar to that of the refrigerants R123 and R245fa and is up to 11 and 50 times lower than with R290 and R744, respectively. The solar absorption chiller using the refrigerant R717 can achieve cooling with a supply temperature up to 5 °C lower than that of R718. The dynamic simulation results show that the energy efficiency of the proposed solar-powered generator–chiller is 14% higher than that of a standard solar-powered absorption chiller. Furthermore, the same solar-powered generator–chiller reduces the primary energy required by a conventional system by 60% (PESr = 0.60). The presented results may be useful for the design of sustainable generator–chillers for rural areas or for autonomous housing in tropical climates.

1. Introduction

In the past, many efforts have aimed to reduce carbon emissions from refrigeration using renewable energy. Due to global warming, efforts have concentrated on power and cooling systems that decrease greenhouse gas emissions. One such study on sustainable cooling in remote areas is presented in [1], which uses wind and solar power to operate a vapor-compression chiller. Their results show that the best performance and return on investment are achieved with ammonia, a refrigerant with low global warming potential. While a solar absorption chiller or an organic solar Rankine cycle, on their own, have limitations such as high initial cost, intermittency, or low efficiency, a solar generator–chiller can offer benefits such as no consumption of fossil fuels and zero polluting emissions. This has been demonstrated by various studies presented below. The study by [2] highlights the need to develop more cost-effective cooling technologies, specifically within the European system. They demonstrated that despite the lower COP of absorption refrigeration, it remains a competitive alternative to vapor-compression refrigeration.

Turja et al. [3] focused on thermodynamic analysis and improvements to integrated energy cycles to increase overall thermal efficiency. They examined advanced CO₂ power cycles and introduced machine learning and genetic algorithms to optimize cycle performance. Their results show a net power output of 7994 MW and a thermal efficiency of 53.36%. That same year, the authors of [4] investigated and evaluated the characteristics and optimal tradeoffs of various cycles with transcritical (t), subcritical (s), and non-azeotropic mixed working fluids. By quantitatively focusing on the sensitivity of target variables within the design space in various combined cycles and visually illustrating the system’s nonlinearity and coupling relationships on 2D weight planes using a Self-Organizing Map (SOM), Moctezuma et al. [5] evaluated the energy and exergy performance of an integrated supercritical CO₂ hybrid Brayton–ORC system with a flat-plate solar collector to improve its efficiency and reduce fuel consumption. They highlighted the fundamental role played by the cycle’s working fluid in the system’s performance.

Studies such as [6,7,8,9] have yielded findings about chillers utilizing various refrigerants and solar energy. Tsoutsos et al. [10] studied the performance and economic evaluation of a solar heating and cooling system for a hospital in Crete. They simulated a complete system consisting of a solar collector, a storage tank, a backup heat source, a water-cooling tower, and a LiBr–H₂O absorption chiller. They observed that the initial investment cost is quite high. However, this is offset by the greater environmental benefits, shorter payback period, and higher total annual savings. As shown in [11,12], the ammonia–lithium nitrate working pair is a promising solution because it can operate at evaporation temperatures below 0 °C and does not require a rectifier.

Systems for power generation and cooling have already been presented in [13,14], where the authors primarily use solutions such as water–lithium bromide and ammonia–water. These refrigerant solutions can operate only at temperatures above zero degrees Celsius and require a rectifier, respectively. The performance evaluation of a cooling plant combining free cooling and a single-effect water–lithium bromide absorption chiller, which has been operating for five years and uses solar energy, was analyzed by [15]. They indicate that truly environmentally efficient systems need to refine their overall design. They demonstrate that in some months of cooling, the energy used for cooling can reach almost 70%. Ma et al. [16] suggested an innovative recompressed supercritical CO₂ Brayton cycle, complemented by an absorption chiller (RSBC/AC), for air-cooled concentrated solar power (CSP) installations. A detailed parametric analysis and optimization demonstrate that the optimal values of PR and dX are obtained using the RSBC/AC to maximize thermal and exergy efficiencies. Similarly, they indicate that because the absorption chillers reduce the TMC, the compressor’s specific work decreases, thereby increasing the net specific power output. Ehyaei et al. [17] conducted a study on a hybrid combined cooling and power (CCP) system powered by geothermal energy, using energy, exergy, and economic analyses. They indicated that applying a LiBr absorption chiller downstream of the organic Rankine cycle (ORC) increased the system’s energy efficiency from 9.3% to 47.3%. However, the exergy efficiency decreased from 15.6% to 4.6%, mainly due to increased exergy destruction in the system. El-Sattar et al. [18] noted that it is possible to model a trigeneration system to convert biomass into electricity, cooling, and heating, with a positive correlation between ammonia concentration and CP, and a negative correlation between ammonia concentration and CR. Furthermore, they report that CPR decreases drastically as the ORC_Pump increases, indicating a decrease in cooling capacity. Recently, ref. [19] studied the feasibility of harnessing solar energy to establish an integrated multi-energy production system to supply heating, cooling, and electricity to passenger trains. They showed that promising performance can meet the energy demands of passenger trains during most operating periods, while producing a surplus of electricity, cooling, and heating that could be stored. Srinivas and Vignesh [20] evaluated the performance of a combined-cycle power plant with inlet air cooling, using a water–lithium bromide absorption refrigeration system powered by waste heat from a heat recovery steam generator (HRSG). They demonstrated that, under certain conditions, combined-cycle output increased by 15% and efficiency by 12% by incorporating vapor absorption refrigeration (VAR) into the plant. In 2017, ref. [21] proposed a combined cycle for cooling, heating, and electricity generation. Their results show that adding a regenerative organic Rankine cycle to a gas turbine and steam generator cycle with heat recovery increases efficiency by 2.5%. Furthermore, adding an absorption refrigeration cycle to the gas turbine and steam generator cycle with heat recovery/regenerative organic Rankine cycle leads to an additional 0.75% increase in the overall cycle’s exergy efficiency.

Based on these findings, conventional refrigerants, such as R11, hydrocarbons, and carbon dioxide, among others, have been used in power cycles coupled to cooling systems. However, only water–lithium bromide and ammonia–water solutions have been used in absorption refrigeration systems coupled with Rankine cycles. This article proposes the following hypothesis: a solar generator–chiller using fluids with low global warming potential to produce electricity and cooling will reduce energy consumption by at least 10% compared to current systems. For this purpose, the performance of a solar generator–chiller with an absorption cycle and a Rankine cycle, using refrigerants with low global warming potential, is analyzed, evaluating each cycle independently and the whole system together. The system is intended for power and cooling generation for off-grid housing in rural areas with tropical climates.

2. Modeling and Simulating

This section presents the modeling and simulation of a generator and a cooler integrated into a solar collector. The modeling is presented separately for each subsystem: the organic Rankine cycle (ORC), the absorption refrigeration cycle (ARC), and the solar heating cycle (SHC). The integrated system is shown in Figure 1. (A) The Rankine cycle in the figure operates as follows: The working fluid is evaporated in the generator using heat from the solar concentrator. The steam produced is fed to the turbine at state 1 to generate mechanical power and exits at state 2 as saturated vapor. The saturated vapor is cooled and condensed into a liquid in the condenser. The pump draws it from the liquid at state 3 and sends it to the generator, which repeats the cycle. (B) The absorption refrigeration cycle operates as follows: The refrigerant at state 3 is a liquid–vapor mixture and is introduced into the evaporator to remove heat from the space to be conditioned, exiting as a refrigerant gas at state 4. The vapor is absorbed by the concentrated solution in the absorber. A pump sends the dilute solution to the generator, passing through the recuperator for preheating. In the generator, the refrigerant is evaporated to state 1 using heat from the solar collector. The refrigerant vapor from state 1 is converted to liquid in the condenser, and the refrigerant mixture, via the expansion valve, repeats the refrigerant cycle. The solution leaving the generator (8) returns to the absorber after losing heat in the recuperator (from 8 to 9) and expanding in the liquid expansion valve (from 9 to 10). (C) The solar heating cycle proceeds as follows: A pump sends the liquid from state (4) to the absorber of the parabolic dish collector, where it exits at a higher temperature in state (2). From state (2) to state (3), the heating fluid transfers heat to the Rankine cycle. From state (3) to state (4), the heating fluid provides thermal energy to the absorption refrigeration cycle. From here on, the cycle repeats. The modeling, validation, and simulation are presented below.

2.1. Modeling

This section presents the modeling of a solar-powered generator–chiller. The general mass and energy balance equations are as follows:

$$\sum {\dot{m}}_i=\sum {\dot{m}}_o,$$

(1)

$$\sum \dot{Q}-\sum \dot{W}=\sum {\dot{m}}_o{h}_o-\sum {\dot{m}}_i{h}_i,$$

(2)

where $\dot{m}$ and h are the mass flow rate and enthalpy, respectively. The subscripts i and o refer to the inlet and outlet, respectively. $\dot{Q}$ and $\dot{W}$ represent heat flow and mechanical power, respectively.

2.1.1. Organic Rankine Cycle Model, ORC

The Rankine cycle was modeled using the energy and mass balances for each component, as described in [4,22].

The heat flow required in the Rankine cycle generator, ${\dot{Q}}_{rg}$, is given by the following:

$${\dot{Q}}_{rg}={\dot{m}}_{rr}(h_1-h_4),$$

(3)

where rr refers to the Rankine cycle coolant, and numbers 1 and 4 are the cycle states.

The turbine power, ${\dot{W}}_T$, is determined from the following:

$${\dot{W}}_T={\dot{m}}_{rr}(h_1-h_2),$$

(4)

The heat flow released in the condenser, ${\dot{Q}}_{rc}$, is determined from the following:

$${\dot{Q}}_{rc}={\dot{m}}_{rr}(h_2-h_3),$$

(5)

The power required by the pump, ${\dot{W}}_{rp}$, is determined from the following equation:

$${\dot{W}}_{rp}={\dot{m}}_{rr}(h_4-h_3),$$

(6)

The thermal efficiency of the Rankine cycle, ${\eta }_r$, is determined from the following equation:

$${\eta }_r={\displaystyle \frac{{\dot{W}}_T}{{\dot{Q}}_{rg}}},$$

(7)

2.1.2. Absorption Refrigeration Cycle Modeling

The modeling of the absorption refrigeration system, ARC, was performed using energy and mass balances, as described in [11,23,24,25], as follows.

The refrigerant flow rate, ${\dot{m}}_r$, is determined from

$${\dot{Q}}_e={\dot{m}}_r\left(h_4-h_3\right),$$

(8)

where $h_4$ and $h_3$ are the inlet and outlet enthalpies inside the evaporator. ${\dot{Q}}_e$ is the heat flow in the evaporator.

The circulation ratio, $f$, which indicates the solution flow required to produce one kg/s of steam, can be determined, as shown in [11,25], in the following form:

$$f={\displaystyle \frac{{\dot{m}}_{ds}}{{\dot{m}}_r}}={\displaystyle \frac{X_{cs}}{X_{ds}-X_{cs}}},$$

(9)

where $X_{ds}$ and $X_{cs}$ are determined from the following:

$$X_{ds}=f\left(P_l,T_5\right),$$

(10)

$$X_{cs}=f(P_h,T_8),$$

(11)

The change in concentration $∆X$ is as follows:

$$∆X=X_{ds}-X_{cs},$$

(12)

Here, $P_h$ and $P_l$ are the low and high pressures, respectively, and are determined from the thermodynamic properties of the refrigerant, along with the evaporator temperature, T_e, and the condenser temperature, T_c.

From this equation, the flow rate of the dilute solution, ${\dot{m}}_{ds}$, can be determined as follows:

$${\dot{m}}_{ds}={\dot{m}}_r{\displaystyle \frac{X_{cs}}{X_{ds}-X_{cs}}},$$

(13)

The flow rate of the concentrated solution, ${\dot{m}}_{cs}$, is determined by mass balance in the absorber, as follows:

$${\dot{m}}_{ds}={\dot{m}}_{cs}+{\dot{m}}_r,$$

(14)

The pump power, ${\dot{W}}_{ap}$, can be determined by changing pressures as follows:

$${\dot{W}}_{ap}={\displaystyle \frac{(P_h-P_l)}{{\eta }_m}}\ast \dot{m_{ds}}\ast v_5,$$

(15)

Furthermore, it can be determined as follows:

$$w_{ap}={\displaystyle \frac{{\dot{W}}_{ap}}{{\dot{m}}_{ds}}},$$

(16)

Therefore, the enthalpy at the pump outlet, $h_6$, can be determined from the following:

$$w_{ap}=(h_6-h_5),$$

(17)

The heat flow transferred in the recuperator, ${\dot{Q}}_r$, is represented as follows:

$${\dot{Q}}_r={\dot{m}}_{ds}(h_7-h_6),$$

(18)

$${\dot{Q}}_r={\dot{m}}_{cs}(h_8-h_9),$$

(19)

The thermal efficiency of the heat exchanger, ${\epsilon }_r$, if $C_{min}$ = $C_h$, is determined as described in [26], using the following equation:

$${\epsilon }_r={\displaystyle \frac{T_7-T_6}{T_8-T_6}},$$

(20)

If $C_{min}$ = $C_c$, the efficiency is calculated as follows:

$${\epsilon }_r={\displaystyle \frac{T_8-T_9}{T_8-T_6}},$$

(21)

where $C_c$ and $C_h$ are the thermal capacitance of the fluids, cold or hot, respectively, and are given as follows:

$$C_c={\dot{m}}_c{Cp}_c,$$

(22)

$$C_h={\dot{m}}_h{Cp}_h,$$

(23)

The heat flows in the components are determined by energy and mass balance and are as follows:

Heat flow in the generator, ${\dot{Q}}_{ag}$, is determined by the following:

$${\dot{Q}}_{ag}={\dot{m}}_r{h}_1+{\dot{m}}_{cs}{h}_8-{\dot{m}}_{ds}{h}_7,$$

(24)

Heat flow in the absorber, ${\dot{Q}}_a$, is determined by the following:

$${\dot{Q}}_a={\dot{m}}_r{h}_4+{\dot{m}}_{cs}{h}_{10}-{\dot{m}}_{ds}{h}_5,$$

(25)

Heat flow in the condenser, ${\dot{Q}}_{ac}$, is determined by the following:

$${\dot{Q}}_{ac}={\dot{m}}_r(h_1-h_2),$$

(26)

The coefficient of performance, $COP$, is determined by the following:

$$COP={\displaystyle \frac{{\dot{Q}}_e}{{\dot{Q}}_g+{\dot{W}}_{ap}}},$$

(27)

2.1.3. Solar Heating Cycle (SHC) Modeling

For modeling the solar heating cycle for a parabolic dish solar collector, the thermal efficiency of the collector is determined as described in [24,27], by using the following:

$${\eta }_c={\displaystyle \frac{\dot{Q_u}}{I{A}_a}}={\eta }_o-{\displaystyle \frac{1}{IC}}[h\left(T_H-T_o\right)+\epsilon \sigma \left(T_H^4-T_o^4\right)],$$

(28)

where I and $A_a$ represent the solar radiation and the collector’s absorber area, respectively. ${\eta }_o$, $C$, and h are, respectively, the optical efficiency, concentration ratio, and convection loss coefficient. ε and σ are the collector emissivity and the Stefan–Boltzmann constant, respectively. $T_h$ and $T_o$ are the temperature of the hot reservoir or absorber and the ambient temperature, respectively.

The useful heat of the parabolic dish solar collector, ($\dot{Q_u}$), is obtained from the following:

$${\dot{Q}}_u=I{A}_a{\eta }_0-A_a\left[h\left(T_H-T_o\right)+\epsilon \sigma \left(T_H^4-T_o^4\right)\right],$$

(29)

The concentration ratio $C$ is determined as follows:

$$C={\displaystyle \frac{A_c}{A_a}},$$

(30)

where $A_c$ and $A_a$ are the collector and absorber areas, respectively.

2.1.4. Sustainable Solar Generator–Chiller Modeling

The modeling of an integrated generator-cooler system powered by solar energy was carried out as described in Section 2.1.1, Section 2.1.2 and Section 2.1.3, for each subsystem. Through energy balance analysis of each system, the following is concluded:

$${\dot{Q}}_u={\dot{Q}}_{ag}+{\dot{Q}}_{rg},$$

(31)

Energy efficiency has been determined as described in [28]. The energy efficiency of the integrated system, for the power cycle with the collector, has been defined as follows:

$${\epsilon }_{rc}={\displaystyle \frac{{\dot{W}}_T-{\dot{W}}_{rp}}{I{A}_c}},$$

(32)

For the absorption refrigeration cycle and the collector, it has been defined by the following:

$${\epsilon }_{ac}={\displaystyle \frac{{\dot{Q}}_e}{I{A}_c}},$$

(33)

Likewise, for the integrated generator-cooler system, it has been defined as follows:

$${\epsilon }_{rac}={\displaystyle \frac{{\dot{Q}}_e+{\dot{W}}_T-{\dot{W}}_{rp}}{I{A}_c}},$$

(34)

2.2. Model Validation

The model presented in Section 2.1 has been validated using results from the literature, as shown in

Table 1. Validation results.

Parameter	Value	This Work	Error, %
Organic Rankine cycle efficiency [4], ${\eta }_r$%	2.7	2.747	1.71
Coefficient of operation, absorption cycle [11], COP	0.78	0.79	1.26
Solar collector efficiency [12], ${\eta }_c$%	0.75	0.82	8

. For the validation of each cycle, the assigned parameters for the Rankine cycle are $P_r$ = 3.84, $T_1$ = 102.3 °C, and $T_c$ = 27 °C. The parameters for the absorption and solar cycles are $T_g$ = 74 °C, $T_{ac}$ = 30 °C, $T_e$ = 0 °C, I = 1000 W/m², C = 100, and $T_H$ = 200 °C. As can be seen, the largest error is 8% for the solar cycle. For the Rankine and absorption cycles, the errors are 1.71% and 1.26%, respectively. Validation has confirmed that when the ambient temperature increases by 3 °C, for example, the thermal efficiency of the Rankine cycle decreases by 2.96%, while when the supply temperature, T_r1, increases by the same proportion, efficiency increases by 2.93%. This suggests that a higher supply temperature is associated with better cycle efficiency. Conversely, the lower the supply temperature, the lower the efficiency.

The specifications for the power cycle, absorption cycle, and solar cycle are presented in

Table 2. Thermal cycle specifications.

Parameter	Value
Organic Rankine cycle
Working fluid	R1233zd(E)
$\mathrm{Turbine} \mathrm{inlet} \mathrm{temperature}, T_{rg}$ °C	89.5
$\mathrm{Condensation} \mathrm{temperature}, T_{rc}$ °C	30
$\mathrm{Pressure} \mathrm{ratio}, P_r$	4.5
Absorption refrigeration cycle
Working fluid	Ammonia–lithium nitrate
$\mathrm{Generator} \mathrm{temperature}, T_{ag}$ °C	74
$\mathrm{Evaporator} \mathrm{temperature}, T_e$ °C	0
$\mathrm{Condensation} \mathrm{temperature}, T_{ac}$ °C	30
Solar heating cycle
Working fluid	Oil Diatermo S
$\mathrm{Collector} \mathrm{temperature} \mathrm{at} \mathrm{outlet}, T_o$, °C	200
$\mathrm{Collector} \mathrm{temperature} \mathrm{at} \mathrm{inlet}, T_i$ °C	100
Solar radiation, I W/m2	1000

.

2.3. Simulating

A simulation of the solar generator–chiller was carried out in steady-state and dynamic modes, as described in this section.

2.3.1. Assumptions

The assumptions for the simulation were as follows:

• The properties of fluids depend on the average temperature during the process;
• The inlet temperature is the same as the outlet temperature in both evaporators and condensers;
• Expansion valves perform an isenthalpic process;
• The condenser and the absorber release heat at the same ambient temperature;
• The simulated ambient temperature and solar radiation values are typical of tropical climates.

2.3.2. Steady-State Simulation

The model in Section 2.1, validated as shown in Section 2.2, was programmed in the specialized software Engineering Equation Solver (EES), V10.228, which identifies and solves a set of equations and simultaneous operations [29]. The thermodynamic properties of the lithium nitrate–ammonia solution were determined using the correlations shown in [30,31]. These have already been used for temperatures greater than 150 °C, among others, as described in [32,33,34,35]. Meanwhile, the properties of the lithium bromide–water solution and of the fluids evaluated with the organic Rankine cycle were obtained from the same specialized software and were provided by the references listed in

Table 3. Thermodynamic properties of organic fluids for power and refrigeration cycles [29].

Trade Name	Nomenclature	Boiling Temperature [°C]	Critic Temperature [°C]	Critic Pressure [kPa]
Genetron–123	R123	27.79	183.7	3668
ENOVATE–245fa	R245fa	15.19	154.0	3651
Eco Flush HFO-1233zd	R1233zd(E)	17.92	109.4	3632
Propane	R290	−42.09	96.68	4247
Carbon dioxide	R744	−78.5	30.98	7377
Water	R718	100	374.0	22,064
Ammonia	R717	−33.32	132.3	11,333

.

The simulation was developed under steady-state conditions to identify working fluids with low global warming potential for use in a low-capacity solar generator-cooler. For the steady-state simulation, three cases were considered: (a) an organic Rankine cycle, (b) an absorption refrigeration cycle, and (c) a generator–chiller using fluids with low global warming potential; each is described below. (A) Rankine cycle: Using the electrical power required by the dwelling (equal to the mechanical power of the turbine) and the pressure ratio as the input variable, the mass flow rate of the working fluid, the pump power, the thermal power of the generator, and the condenser power are determined. (B) Refrigeration cycle: Using the data for the cooling capacity required by the dwelling, as well as the evaporation, condensation, and generation temperatures, the refrigerant flow rate, the solution flow rate, and the thermal power of the generator, condenser, and absorber are determined. (C) Collector cycle: Using the data for solar radiation and ambient temperature, as well as assigning an operating temperature for the absorber, the efficiency of the parabolic dish collector, the useful heat output, and the collector area are determined. (D) Solar generator–chiller: With a given evaporation temperature and solar radiation, the thermal power and thermal efficiency of the power cycle, the collector area and efficiency of the cooling cycle, and the energy efficiency of the coupled system were determined for different supply and condensation temperatures. The simulation of the coupled system was carried out as shown in the flow diagram in Figure 2. The figure shows that the input data are the electrical power, the cooling capacity, and the required dwelling temperature. The process begins by selecting the working fluid. First, the fluid for the power cycle is evaluated: organic, natural, and/or hydrocarbon. After assigning climatological conditions, such as solar radiation and ambient temperature, the refrigerants for the cooling cycle are evaluated: R718 and R717. The thermal and mechanical power outputs, the collector area, and the coupled system’s efficiencies are then determined.

2.3.3. Dynamic Simulation

In the dynamic simulation, in addition to analyzing the system using climatological data, the generator and the chiller were compared separately with conventional units. The specifications of the reference equipment are presented in

Table 4. Specifications of the conventional generator–chiller.

Parameter	Value
Conventional generator
$\mathrm{Thermal} \mathrm{efficiency}, {\eta }_t$	0.30
$\mathrm{Electrical} \mathrm{power} \mathrm{demand}, {\dot{W}}_{T,D}$ kW	1.5
Conventional chiller
$\mathrm{Cooling} \mathrm{power} \mathrm{demand}, {\dot{Q}}_{eD}$, kW	3.517
$\mathrm{Coefficient} \mathrm{of} \mathrm{performance}, {COP}_{ch}$	2.5
$\mathrm{Thermal} \mathrm{efficiency} \mathrm{of} \mathrm{the} \mathrm{chiller} \mathrm{source}, {\eta }_t$	0.30

.

The thermal power required by the conventional generator was determined as follows. From the thermal efficiency of a heat engine, ${\eta }_t$, the heat supplied, ${\dot{Q}}_{rD}$, is determined by the following:

$${\eta }_t={\displaystyle \frac{{\dot{W}}_N}{{\dot{Q}}_{rD}}},$$

(35)

where ${\dot{W}}_N$ is the net power produced by the heat engine. In this study, ${\dot{W}}_N$ is the electrical power demanded by the house.

The thermal power required by a conventional air conditioning unit was determined as follows. The coefficient of performance (COP) equation is given by thew following:

$$CO{P}_c={\displaystyle \frac{{\dot{Q}}_e}{{\dot{W}}_c}},$$

(36)

If the mechanical power of the compressor is considered equal to the net power of the heat engine, ${\dot{W}}_c={\dot{W}}_N$, then, according to Equation (35), the compressor power is as follows:

$${\dot{W}}_c={{\eta }_t\dot{Q}}_s,$$

(37)

Substituting ${\dot{W}}_c$ in Equation (36) and solving for ${\dot{Q}}_{aD}$, the thermal power demanded by a conventional air conditioning unit is given by the following:

$${\dot{Q}}_{aD}={\displaystyle \frac{{\dot{Q}}_{eD}}{CO{P}_c{\eta }_t}},$$

(38)

In this work, ${\dot{Q}}_{eD}$ is the cooling capacity demanded by the dwelling.

Energy profitability is determined separately for both the generator and the chiller using primary energy saving (PES) and the primary energy saving rate (PESr), as described in [36,37]. For the Rankine generator, they are determined by the following:

$$PES={\dot{Q}}_{rD}-{\dot{Q}}_{rg},$$

(39)

$$PESr={\displaystyle \frac{{\dot{Q}}_{rD}-{\dot{Q}}_{rg}}{{\dot{Q}}_{rD}}},$$

(40)

However, for the absorption chiller, they are given by the following:

$$PES={\dot{Q}}_{aD}-{\dot{Q}}_g,$$

(41)

$$PESr={\displaystyle \frac{{\dot{Q}}_{aD}-{\dot{Q}}_g}{{\dot{Q}}_{aD}}},$$

(42)

3. Results

This section presents the simulation results for identifying working fluids with low global warming potential and for evaluating the performance of a solar generator-cooler system for homes in rural or off-grid areas in tropical climates. First, the performance of the Rankine cycle is evaluated; then, that of an absorption chiller; and finally, that of an integrated solar generator–chiller system, both under steady-state conditions and in dynamic simulation.

3.1. Low Global Warming Potential Power Cycle Assessment

To evaluate the performance of the Rankine cycle using low-global-warming-potential fluids, refrigerants R123, R245fa, and R1233zd(E) were used, along with hydrocarbons such as R290 (propane) and natural gas such as R744 (carbon dioxide). Figure 3 shows the overall behavior of a Rankine cycle at low pressures ($P_l$) and high pressures ($P_h$) with respect to the pressure ratio ($P_r=P_h/P_l$) for the Rankine cycle using the five working fluids. The figure shows that refrigerant R1233zd(E) can operate at around 170 kPa on the low side and 900 kPa on the high side, with a pressure ratio of six. These pressures are similar to those of current systems that use refrigerants R123 and R245fa. However, using hydrocarbons R290 and R744 at the same pressure ratio, the operating pressures are 11 and 50 times higher (for R290 and R744, respectively) than those of R1233zd(E). It can be concluded that if R1233zd(E) is used as the working fluid instead of R290 or R744, operating pressures can be up to 11 or 50 times lower, thereby reducing installation costs.

Figure 4 shows the turbine inlet temperature as a function of the pressure ratio for the five evaluated refrigerants. The figure indicates that with refrigerant R1233zd(E), turbine inlet temperatures can range from approximately 52 °C to 118 °C. This means that the Rankine cycle can produce power at approximately 85 °C with a pressure ratio of six. This is up to 30% lower operating temperatures than with hydrocarbons such as R290 or R744, which require approximately 120 °C at a pressure ratio of six.

Figure 5 shows the heat flow required by the generator and released by the condenser, as a function of the pressure ratio, for the five evaluated fluids. From the figure, at pressure ratios of two and three, the heat flow required for cycle operation is up to 2.5 and 1.5 times higher, respectively, than at a ratio of six. Similarly, at a pressure ratio of six, the heat flow required for R744 is up to 85% lower than for R1233zd(E). The heat flow released in the condenser is in a similar proportion. From this figure, it can be concluded that the heat flow required by a Rankine cycle operating with R1233zd(E) is stable at pressure ratios between five and seven, and is up to 150% lower than two and three, respectively.

Finally, Figure 6 presents the thermal efficiency of the Rankine cycle as a function of the pressure ratio for the five evaluated working fluids. The figure shows that the efficiency of R1233zd(E) is similar to that of the other organic fluids (R123 and R245fa), and even to that of the hydrocarbon R290, for example, at a pressure ratio of six. The results show that R1233zd(E) can be used with pressure ratios of six and efficiencies of 14%.

3.2. Low Global Warming Potential Absorption Chiller Assessment

This section evaluates the influence of condensation and generator temperatures on the performance of a solar absorption chiller operating with natural refrigerants of low global warming potential. The evaluated working fluids were water–lithium bromide (H₂O–LiBr) and ammonia–lithium nitrate (NH₃–LiNO₃), which are based on the natural refrigerants R718 (water) and R717 (ammonia), respectively.

The evaluation was performed considering the cooling capacity of one ton of refrigeration, ${\dot{Q}}_e$, (3.517 kW), I = 600 W/m², and $T_e$ = 4 °C. Figure 7 shows the change in concentration in the generator with respect to the generator’s temperature at three condensation temperatures for the two operating pairs, R718 and R717. The dashed lines included in Figure 7, Figure 8, Figure 9, Figure 10 and Figure 11 indicate the operating limit of the systems, according to References [11,25,38], that is, the minimum generator temperature required to achieve a sufficient concentration change for effective refrigerant separation. Once the start-up threshold is exceeded, the concentration change increases with the generator temperature, reflecting an intensification of the separation process and a greater cooling capacity. The NH₃–LiNO₃ pair achieves higher ΔX values than the H₂O–LiBr pair, indicating greater mass transfer efficiency and a higher potential cooling capacity. Furthermore, the NH₃–LiNO₃ pair starts to work at generator temperatures 5 °C lower, making it more suitable for applications with low-quality heat sources, such as solar power. This difference in performance can serve as a criterion for selecting the most appropriate working pair based on the available thermal conditions and the system’s operating requirements.

Figure 8 shows the variation in the heat flow in the generator (${\dot{Q}}_g$) as a function of the generator temperature for two absorption refrigeration systems using the H₂O–LiBr and NH₃–LiNO₃ working pairs. Three condensation temperature scenarios are analyzed, 24 °C, 30 °C, and 36 °C, simulating conditions for tropical climates. The results show that, for both working solutions, the heat flow in the generator decreases with increasing generator temperature. The NH₃–LiNO₃ pair exhibits higher ${\dot{Q}}_g$ values, indicating greater thermal energy absorption once it begins to operate. However, this NH₃–LiNO₃ solution requires generator temperatures lower than 5 °C, compared to the H₂O–LiBr solution, to initiate the cycle, making it more suitable for applications with low-quality heat sources. This difference in performance allows for optimization of the selection of the working fluid based on the available thermal conditions.

Figure 9 compares the performance of two absorption refrigeration systems using the working solutions H₂O–LiBr and NH₃–LiNO₃ at three different condensation temperatures (24 °C, 30 °C, and 36 °C) and with respect to the generation temperature. The behavior of the working pairs shows significant differences: NH₃–LiNO₃ begins its useful operation at a generator temperature of 5 °C or lower, making it more suitable for low-quality heat sources. Furthermore, it maintains more stable performance despite variations in condensation temperature. In contrast, H₂O–LiBr requires higher temperatures to overcome the initial operating temperature. This suggests that NH₃–LiNO₃ is more variable under tropical environmental conditions.

Figure 10 shows the variation in the required solar collector area as a function of the generator temperature for absorption refrigeration systems using H₂O–LiBr and NH₃–LiNO₃ working solutions, evaluated at three different condensation temperatures: 24 °C, 30 °C, and 36 °C. From the start of cycle operation (intersection with the dotted lines) and up to 2 °C or 6 °C after the start of operation, the solar collector area decreases as the generator temperature increases. However, after 2 °C or 6 °C after the start of the absorption cycle, the collector area increases at a rate of 0.0893 m²/°C (893 cm²/°C) and 0.0765 m²/°C (765 cm²/°C) for the water–lithium bromide and ammonia–lithium nitrate chillers, respectively. This is obtained between 71.2 °C and 94.7 °C in the generator and $T_c$ = 24 °C. If we analyze, for example, a temperature of 73.6 °C in the generator with $T_c$ = 24 °C, we can see that the NH₃–LiNO₃ pair requires a collector area 8.7% larger than when H₂O–LiBr is used. This can increase the acquisition cost of a solar system powering an ammonia–lithium nitrate system for space conditioning applications. However, it resolves the crystallization and vacuum-pressure problems associated with water–lithium bromide systems, as the ammonia–lithium nitrate solution does not exhibit these limitations.

Figure 11 shows the heat flow released in the absorber as a function of the generator temperature for an absorption refrigeration system using the H₂O–LiBr and NH₃–LiNO₃ work pairs at three condensation temperatures: 24 °C, 30 °C, and 36 °C. The figure shows that between 4 and 7 kW of thermal energy is lost to the atmosphere in the absorber for a 1 TR (3.517 kW) absorption chiller at condensation temperatures between 24 °C and 36 °C. This thermal energy could be used for cogeneration or heating systems.

3.3. Low-Global-Warming-Potential Solar Generator–Chiller Steady-State Simulation

This section presents the results of the integrated solar-powered generator–chiller, using natural refrigerants with low global warming potential. The results were obtained for an electrical power output of 2 kW from the generator coupled to the turbine, a pressure ratio of six, and a cooling capacity of 3.517 kW in the evaporator at an evaporator temperature of 0 °C. Figure 12 shows the change in concentration in the absorption chiller with respect to the temperature supplied by the solar collector, T_H, for three condensation temperatures (three ambient temperatures). The figure shows that the change in concentration increases with increasing supply temperature. It can also be seen that as the condensation temperature increases, the concentration change decreases by approximately 15% and 33% at 28 °C and 32 °C, respectively, compared to $T_c$ = 24 °C.

Figure 13 shows the heat flow supplied by the collector, ${\dot{Q}}_u$, to the absorption cycle, ${\dot{Q}}_{ag}$, and to the power cycle, ${\dot{Q}}_{rg}$, with respect to the feed temperature, $T_h$, for three condensation temperatures. It is assumed that the condensation temperature of the absorption cycle is the same as that of the Rankine cycle ($T_{ac}$ = $T_{rc}$). The figure shows that the heat flow for all three cases decreases as the supply temperature increases. It can also be observed that, at the same supply temperature, $T_h$ = 118 °C, the heat flow increases by approximately 7% and 10% for the absorption and Rankine cycles, respectively, as the condensation temperature increases from 24 °C to 32 °C. The results suggest that, if the supply temperature increases from 105 °C to 140 °C, for example, with $T_c$ = 32 °C, the heat flow supplied to the power cycle and the cooling cycle decreases by up to 37% and 3%, respectively.

Figure 14 presents the Rankine cycle efficiency, the absorption cycle coefficient of performance (COP), and the collector efficiency as a function of the supply temperature for three ambient temperatures. The figure shows that the power cycle efficiency and the COP increase with the supply temperature and decrease with the condensation temperature. However, collector efficiency decreases with increasing hot temperature and increases with solar radiation. The figure also shows that the Rankine cycle efficiency exceeds 22% at a feed temperature of 105 °C. Furthermore, the performance absorption cycle coefficient is greater than 0.7 for the entire case studied, as also recommended by [24]. Additionally, the collector efficiency is above 80%, as reported by [12]. These results are favorable for the operation of the integrated system.

Figure 15 shows the collector and absorber areas as a function of the hot-source temperature for three ambient temperatures. As the supply temperature increases, the required collector area decreases. For example, at 120 °C, the collector area is approximately 16.5 m², and 17.5 m² for condensation temperatures of 24 °C and 32 °C, respectively. Based on these collector area values, it can be concluded that this integrated solar system for power production and sustainable cooling can be installed in homes with available areas of at least 18 m² (only one room of a 4 m × 5 m home meets this requirement).

The collector efficiency with respect to the Willier–Bliss–Hottel coefficient, ($T_i-T_o/I)$, for the entire evaluated range, is shown in Figure 16. Efficiency decreases as the Hottel coefficient increases, as ambient temperature rises, and as solar radiation decreases. However, within the radiation range of 600 to 1000 W/m² and ambient temperatures of 24 to 32 °C, the efficiency remains approximately from 80.5% to 84.3%. The range of values obtained for the solar collector’s efficiency supports the use of this solar energy equipment.

Figure 17 shows the energy efficiency of the integrated system as a function of the supply temperature for three ambient temperatures. For example, at 120 °C, the solar efficiency of the Rankine cycle alone is approximately 16%. This efficiency is approximately 25% when the collector is used for the absorption cycle. However, when integrated power and absorption cycles are used, energy efficiency increases to approximately 42% at an ambient temperature of 24 °C. It can be concluded that using parabolic dish solar collectors to produce electricity and cool with a refrigerant of low global warming potential can improve energy efficiency by 17%.

3.4. Low-Global-Warming-Potential Solar Generator–Chiller Dynamic Simulation

The solar-powered generator–chiller described in Section 2 was evaluated and compared to a conventional one using dynamic simulation with climatological data from the tropical city of Villahermosa, Tabasco, Mexico. This location was chosen because, for most of the day and year, the climatological conditions of radiation and temperature are higher than those in regions near the Earth’s poles [39]. Figure 18 presents 8784 data points of solar radiation and ambient temperature for 2024 [40], used for the dynamic simulation of the generator–chiller. The figure shows that the maximum radiation in winter and summer is 700 W/m² and 1000 W/m², respectively, while the ambient temperature in this tropical region reaches a maximum of 40 °C in spring and a minimum of 18 °C in winter. Based on climatological data from this tropical region, it can be concluded that, because the ambient temperature is above 23 °C most of the time, it is necessary to seek energy alternatives to maintain comfortable temperatures and reduce energy consumption.

The solar-powered generator–chiller model shown in Section 2.1 was evaluated in a dynamic simulation using climatological data from 8784 h in 2024. The dynamic simulation is presented for a typical day of each climatological season, for the twelve months of the year, and as energy profitability.

3.4.1. Typical Seasonal Day

Figure 19 shows the electricity and cooling produced by the generator-cooler for a typical day in each season. The selected days were midway through each season: February 8, May 7, August 7, and November 5 for winter, spring, summer, and autumn, respectively. The figure also shows the electrical power (1.5 kW) and cooling capacity (3.517 kW) demanded by the house and provided by conventional equipment. The specifications for this conventional equipment are presented in Table 4. The figure shows that electricity and cooling are required 24 h a day. However, the electricity and cooling provided by the generator-cooler are only representative between approximately 10:00 AM and 6:00 PM, across all four seasons. The generator–chiller provides electricity and cooling for at least 8 h a day. Furthermore, the generator–chiller produces approximately 50% more energy than required for 5 h a day. The resulting improvement from coupling these two systems is approximately 14% compared to the absorption chiller alone.

3.4.2. Energy Efficiency

As described in [28,41], the energy efficiency of a new system over a full year can be expressed in terms of its monthly energy output. Figure 20 shows the electrical power output of the organic generator, the cooling power output of the absorption chiller, the useful heat output of the solar collector, and the thermal power input to the solar-powered generator–chiller. The figure shows that the highest energy output occurs in April, May, and August, while January and December have the lowest. This is because solar radiation is higher in these months and lower in others. The figure also shows that the maximum energy output in the months with the highest radiation is around 600 kWh for mechanical power and 1400 kWh for thermal power. The minimum output is approximately 400 kWh and 750 kWh, respectively.

As described in [41] and according to Equations (32)–(34), the energy efficiency of the generator, the chiller, and the coupled generator–chiller is presented in Figure 21. The figure shows that throughout the year, the energy efficiency of the generator and the chiller separately is approximately 15% and 27%, respectively. Meanwhile, the efficiency of the coupled generator–chiller is approximately 41%. The improvement from coupling these two systems is approximately 14% compared to the absorption chiller alone.

3.4.3. Energy Profitability

Energy profitability is determined as described in Section 2.3.3, taken from [36,37], using primary energy consumption. The primary energy consumed annually by a conventional generator and chiller, as well as by the generator and chiller proposed here, is shown in Figure 22a and Figure 22b, respectively. The figures show that the peak primary energy consumption of the organic generator and the absorption chiller is approximately 1700 kWh in April, May, and August. The minimum primary energy consumption of 1000 kWh occurs in January and December.

From the results obtained using Equations (39)–(42) [36,37], the primary energy saving (PES) for the organic generator and absorption chiller are 26,056 MWh and 23,334 MWh, respectively. Meanwhile, the primary energy saving rate (PESr) for the generator and the cooler are 0.60 and 0.57, respectively. If energy profitability is calculated using the conventional generator and chiller together and with the solar-powered generator–chiller coupled, the primary energy savings and the primary energy savings rate are PES = 49.390 MWh and PESr = 0.60. This PESr value of 0.60 can be interpreted as a 60% primary energy saving by the solar generator–chiller proposed here compared to a conventional one.

4. Conclusions

The results obtained in this work are presented for both steady-state and dynamic simulations. The Rankine cycle using R1233zd(E) could reduce installation costs, as operating pressures are up to 11 or 50 times lower than with R290 or R744. This cycle, with a pressure ratio of six, begins to produce power at 85 °C, unlike R290 or R744, which start at 120 °C. Operation of the Rankine cycle is recommended at pressure ratios of 5–7.

The NH₃–LiNO₃ pair achieves higher ΔX values than H₂O–LiBr, indicating greater mass transfer efficiency and a higher potential cooling capacity. It was observed that with this pair, the operation starts 5 °C lower than with water–lithium bromide. This makes it more suitable for applications with low-quality heat sources, such as solar energy. When using the NH₃–LiNO₃ pair, the solar collector area is approximately 8.7% larger than with H₂O–LiBr. However, the advantage of using the ammonia–lithium nitrate solution is that it operates at positive pressures and does not exhibit crystallization, unlike those using the water–lithium bromide pair.

The solar-powered generator–chiller can produce electricity and cooling for homes with available surfaces of 18 m², which is provided by one room (4 m × 5 m = 20 m²). The results suggest that if the supply temperature increases from 105 °C to 140 °C, for example, with $T_c$ = 32 °C, the heat flow supplied to the power cycle and the cooling cycle decrease by up to 37% and 3%, respectively. Similarly, for example, at a supply temperature of 120 °C and an ambient temperature of 24 °C, the solar efficiency of the generator–chiller is 42%. This is 17% higher than that of the cooler alone.

In tropical climates, ambient temperatures exceed 23 °C most of the time, making alternative energy sources essential to reduce energy consumption and maintain comfortable temperatures. The solar generator–chiller provides electricity and cooling for at least 8 h a day. Furthermore, for approximately 5 h a day, the generator–chiller produces up to 50% more energy than a home requires. The combination of these two systems improves energy efficiency by approximately 14% compared to an absorption chiller alone.

The primary energy savings (PESs) for the organic generator and absorption chiller are 26,056 MWh and 23,334 MWh, respectively. Meanwhile, the primary energy saving rates (PESrs) for the generator and the cooler are 0.60 and 0.57, respectively. The primary energy saving and primary energy saving rate of the solar-powered generator–chiller are PES = 49.390 MWh and PESr = 0.60, respectively. The results show that the solar generator–chiller reduces the primary energy required by a conventional one by 60% (PESr = 0.60).

Future work is recommended to develop experiments under real operating conditions to validate simulation results and confirm the system’s technical feasibility in residential environments; evaluate the solar generator–chiller under different operating conditions and climatic regions to determine its global applicability; and integrate thermal storage to improve solar energy performance.