TY - JOUR
AU - Agyenim,, Francis
AB - Abstract This study assessed through numerical simulations, the technical feasibility of a solar-powered absorption cooling system for a small-scale application in an office building in three different cities with a tropical climate in Ecuador. The model and simulations were performed using the dynamic transient software TRNSYS and were compared and validated using experimental data obtained from a real-life system with main components: 12 m2 vacuum tube solar thermal collector array, a 4.5 kW LiBr/H2O single-effect absorption chiller, a 6 kW fan coil and a 100 l sensible cold store. The results of the simulation showed a good agreement with the experimental data with a deviation of 8.5%. The validated model was used to undertake a parametric study to determine capacities of systems that will be applicable to three Ecuadorian cities: Guayaquil, Manta and San Cristobal. The system capacity predicted by the model for the Ecuadorian cities has the following components: 24 m2 evacuated tube collector field, a 20 kW heat exchanger, a 15 kW single-stage LiBr/H2O absorption chiller, a 35 kW cooling tower and a cold storage tank of 2 m3. The results showed that the proposed system could meet most of the required cooling load (90% for Guayaquil and San Cristobal, and 71% for Manta, considering a set point of 24°C) of a typical single-story office building with. 1 INTRODUCTION Five of the warmest years on record occurred since 2010; the peak was in 2016 when the global average temperature was 0.56°C above 1981–2010 average [1]. This global warming trend has a significant impact on the energy sector, both in generation and demand. Regarding the demand side, various authors [2,3] have pointed out that the use of air conditioning is increasing worldwide, due not only to higher ambient temperatures but also to economic growth in developing countries. Considering the above, the use of renewable energy sources in powering air conditioners is a valuable strategy to reduce or offset the growth in cooling demand; especially with the use of active solar systems. In solar-powered air conditioning, cooling loads are in phase with the amount of solar radiation. Indeed, the use of solar cooling technologies has the potential for primary energy savings in the range of 21–70% [4]. One of the first solar active cooling systems was a solar absorption air conditioner of industrial-scale constructed and tested in 1976 in the USA [5]. At the present time around 1000 solar thermal cooling systems are installed worldwide [6], the majority of them in Europe. This is partly due to the support of programs such as Solar Heating and Cooling established in 1977 by the International Energy Agency (IEA), and projects supported by the European Commission: Solar Air Conditioning in Europe from 2002 to 2004, and SOLAIR which ended in 2009. Of all the available options to harvest solar energy for cooling production, thermally driven absorption is the leading technology, this is because of the following: Solar thermal collectors have higher efficiency (greater than 70% for vacuum tube collectors) than photovoltaic modules (about 15%); Solar thermal cooling can be integrated into existing domestic hot water—DHW infrastructure in residential and commercial buildings [7]; The use of solar-powered air conditioning reduces the environmental impact related to the use of refrigerants that contain chlorofluorocarbons and hydrochlorofluorocarbons in conventional compressor chillers [8]. Regarding the capacity of active solar cooling systems, the majority of studies and installations currently in operation are large scale; consequently, further research in small solar cooling and air conditioning applications is essential to promote this technology in the market [9–11]. Most of the studies on solar-powered air conditioning have been developed in areas where the cooling demand is seasonal, and therefore standalone solar cooling systems are not economically viable. The potential for economic gains is high in tropical regions where both cooling loads and solar radiation are higher and less variable all year round [12]. Although there are only a few studies developed for those locations [13–21], the general conclusion is that solar cooling applications have good performances in terms of energy savings. Various researchers have highlighted the need for more building integration of renewable energy systems [12,14,22,23] in order to cope with challenges being faced today: resources abatement and climate change. Solar-driven cooling systems are promising both in terms of energy savings and lower environmental impact [24]. To accomplish that, it is required to develop integral dynamic simulations and analysis for the design of systems that can be energy and cost-effective and therefore economically competitive [18]. Considering the above-mentioned the main purpose of this paper is to evaluate the technical feasibility of a small-scale solar-powered absorption cooling system in three Ecuadorian cities with a tropical climate. The three cities selected for the study were Guayaquil, Manta and Puerto Baquerizo Moreno. Guayaquil and Manta share similar topography as well as climate and proximity to the Pacific coast. Puerto Baquerizo Moreno is in San Cristobal, which is part of the Galapagos Islands. The selection of Puerto Baquerizo Moreno was based on the offer of an alternative to the traditional vapor-compression cycle and the recent growth in tourism at the Galapagos Islands [25], which has led to increased pressure on buildings and services as well as the island’s fragile and unique environment. This paper reports on the development and validation of a TRNSYS model, validated using experimental data followed by a parametric study of initial model employing data from the tropical climate of Guayaquil, Manta and Puerto Baquerizo Moreno in Equador. 1.1 Energy context and climate Ecuador is a South American country with an area of 283 561 km2, has an approximate population of 16.7 million and lies between latitude 01°30′N and 03°23.5′S, and longitude 75°12′W and 81°00′W. Its climate is significantly influenced by two ocean currents: the warm current known as ‘El Niño’ that reaches the Ecuadorian coasts on the last days of December, and the cold current of Humboldt coming from the South Pole. As a result, there are two distinctive seasons, and for coastal cities, they are characterized by hot and rainy periods between late December and May. From June to early December the weather is less humid and a few degrees cooler with an average temperature of 25°C on the coastal region. The electricity production in Ecuador has shown a polynomial growth in the past quinquennium, from 11 944 GWh in the year of 2002 to 28 051 GWh in 2017 [26]. This trend is mainly due to the steady increase of not only the industrial sector but also the residential and commercial sectors, which represented 36% and 19%, respectively, of the electric energy demand in 2017 [27]. In the period between 2002 and 2017, electricity consumption in buildings was more than half of the total country’s demand; reaching 55% in 2017 (Figure 1). The preceding is a consequence of both an escalation on the number of residential and commercial users as well as an improvement in living standards of the people, with average income rising from US$233.13 per month in 2008 to US$437.44 in 2017 [28]. Figure 1 Open in new tabDownload slide Electricity consumption by different sectors in Ecuador, period 2000–17 [27]. Figure 1 Open in new tabDownload slide Electricity consumption by different sectors in Ecuador, period 2000–17 [27]. Even though most of the electricity in Ecuador comes from hydropower (72% of the total production in 2017) [26], the continuous growth in air conditioning market dictated the need for an additional peak load increases. Additional supply will not necessarily come from renewable sources. The use of thermal cooling systems, therefore, represents an opportunity to offset the peak loads increase. Annual average global solar radiation in Ecuador ranges from 4.4 to 4.7 kWh/m2/day [29]. By Ecuador’s location, the incident angle of the solar radiation is perpendicular to the surface during the whole year; therefore the annual solar radiation intensity is uniform, reducing the variations and making the use of solar energy technologies more reliable [30]. The average of direct solar radiation along the coast (which includes Guayaquil and Manta) is below 2.4 kWh/m2/day, and the diffuse radiation ranges from 2.6 to 3.1 kWh/m2/day in the months of higher humidity [29], which is the rainy season when the cooling demand is greater. Guayaquil is a tropical climate city located 10 m above sea level on the Pacific Coast and 2° below the Equatorial line, with minimum and maximum mean ambient temperatures of 18°C and 34°C, respectively. The hourly profile of global insolation and ambient temperature are shown in Figure 2. Figure 3 shows average daily insolation and temperature of Manta and Puerto Baquerizo Moreno. Manta is a port city, as Guayaquil, with similar weather and solar irradiation; the mean average temperature is 24.8°C and monthly maximum and minimum temperatures are 30.9°C and 20.3°C, respectively. Puerto Baquerizo Moreno is in San Cristobal, which is part of Galapagos Islands. Its average temperature is in the range of 20–32°C, and the highest temperatures are recorded from late December to May. Figure 2 Open in new tabDownload slide Hourly profile of insolation (kJ/h.m2) and ambient temperature (°C) in Guayaquil, Average year. Figure 2 Open in new tabDownload slide Hourly profile of insolation (kJ/h.m2) and ambient temperature (°C) in Guayaquil, Average year. Figure 3 Open in new tabDownload slide Daily insolation (kJ/h.m2) and temperature (°C) of (a) Manta and (b) Puerto Baquerizo – San Cristobal. Figure 3 Open in new tabDownload slide Daily insolation (kJ/h.m2) and temperature (°C) of (a) Manta and (b) Puerto Baquerizo – San Cristobal. 1.2 Experimental and TRNSYS model set-ups A system-oriented simulation tool was used to assess the technical feasibility of the solar-powered cooling equipment. However, as the performance of an absorption chiller and the conditioned space are mutually influenced by each other, a software capable of simulating the conditioned space in conjunction with the solar-powered cooling system was required. The commercial software TRNSYS was selected. TRNSYS offers a variety of subroutines models, known as types, and allows interconnection among them to represent complex real systems. 1.3 Base model The base model was established from the solar-powered absorption cooling system set-up and tested by Agyenim et al. [31] during the summer of 2007 at Cardiff, UK. This configuration was preferred because absorption technology has better performance among sorption cooling equipment; evacuated tube collectors (ETCs) are more efficient than flat plate collectors (FPCs), and they can harvest more of the incoming solar radiation, both direct and diffuse. the model used only cold water storage to bridge the gap between generation and use when solar is not available. The main advantage is that they can cover demand peaks when using small capacity chillers and/or the daily operation period is long, as it is often the case in residential small-scale applications. In addition, the part-load or intermittent operation is avoided [32]. The proposed configuration for the initial model can be considered as two sub-systems: the solar thermal loop and the cooling loop. The two loops are interconnected to each other by a 40 kW heat exchanger, as it is presented in Figure 4. The solar thermal sub-system includes a 12 m2 Thermomax vacuum tube collector array and one pump; whereas the cooling sub-system consists mainly of a 4.5 kW LiBr/H2O Rotartica semi-commercial single-effect absorption chiller, a 1000 l cold-water storage tank, a 6 kW fan coil unit, pumps and the conditioned space. Figure 4 Open in new tabDownload slide Schematic diagram of the prototype solar absorption cooling system installed and tested during the summer of 2007 at Cardiff, UK [31]. Figure 4 Open in new tabDownload slide Schematic diagram of the prototype solar absorption cooling system installed and tested during the summer of 2007 at Cardiff, UK [31]. Figure 5 Open in new tabDownload slide Configuration of the solar powered absorption cooling system in TRNSYS. Figure 5 Open in new tabDownload slide Configuration of the solar powered absorption cooling system in TRNSYS. The prototype [31], which comprised of four heat transfer loops (the solar loop, the chiller inlet loop between the heat exchanger and the chiller, the chiller output loop between the chiller and the cold store and the fan coil loop), was in operation during summer months (August and September) of 2007. To evaluate the system performance, in each of the four loops, a data-acquisition system was installed, which comprises of 16 thermometers, 4 flow meters and 1 pyranometer. In addition, ambient conditions were measured while testing the prototype. The system performance was evaluated by integrating the difference in energy input and output over time, with measurement time steps of 5 minutes. Figure 5 shows the model of the solar absorption cooling system developed in TRNSYS. The weather data in the model are not corresponding to the period when the testing was executed; in the model a typical meteorological year -version 2- TMY2 weather data file was included. Moreover, the model type used in this study to represent the small-scale chiller was developed in TRNSYS only for water-cooled units [33], which differs from the chiller used in the prototype that was air-cooled. Finally, in this model, the effects of inertia of the equipment were not included. Below is the description of each of the components of the model. The simulations of the ETC array were performed using TRNSYS Type 71 routine, which determines the thermal performance of the total collector array using the steady-state quadratic efficiency equation, known as Hottel-Whillier equation [34]. $$\begin{equation}\eta ={\eta}_o-{a}_1\left(\frac{\Delta T}{I_T}\right)-{a}_2\left(\frac{\Delta{T}^2}{I_T}\right)\end{equation}$$ (1) The model of the ETC was defined based on the technical specifications of the collector DF100 by Thermomax, which are shown in Table 1. The proposed system has a heat exchanger to transfer the thermal energy from the working fluid in the collector loop, Tyfocor, to the chiller working fluid, water. The heat exchanger was included in the model using Type 91, which simulates the performance of a zero capacitance sensible heat exchanger with constant effectiveness, independent of the system configuration [34]. The parameters of this model are the effectiveness of the heat exchanger, 0.9, and the specific heat of fluids, 3.90 kJ/kg-K and 4.19 kJ/kg-K for Tyfocor and water, respectively. Table 1 Technical specifications of ETC. Parameter Unit Value Collector type - Direct flow vacuum Number of tubes - 30 Dimensions–gross- mm 1996 × 2127 × 97 Absorber area m2 3.020 Weight–empty- kg 81.4 Fluid content Lt 5.6 Max. operating pressure bar 8 Flow rate -nominal kg/h-m2 80 Efficiency: based on absorber area ƞo - 0.832 a1 kJ/hr-m2-K 4.104 a2 kJ/hr-m2-K2 0.0518 Fluid specific heat: Tyfocor kJ/kg-K 3.6–3.9 Fluid density: Tyfocor kg/m3 980–1035 Max. fluid temperature °C 170 Parameter Unit Value Collector type - Direct flow vacuum Number of tubes - 30 Dimensions–gross- mm 1996 × 2127 × 97 Absorber area m2 3.020 Weight–empty- kg 81.4 Fluid content Lt 5.6 Max. operating pressure bar 8 Flow rate -nominal kg/h-m2 80 Efficiency: based on absorber area ƞo - 0.832 a1 kJ/hr-m2-K 4.104 a2 kJ/hr-m2-K2 0.0518 Fluid specific heat: Tyfocor kJ/kg-K 3.6–3.9 Fluid density: Tyfocor kg/m3 980–1035 Max. fluid temperature °C 170 Source: [35] Open in new tab Table 1 Technical specifications of ETC. Parameter Unit Value Collector type - Direct flow vacuum Number of tubes - 30 Dimensions–gross- mm 1996 × 2127 × 97 Absorber area m2 3.020 Weight–empty- kg 81.4 Fluid content Lt 5.6 Max. operating pressure bar 8 Flow rate -nominal kg/h-m2 80 Efficiency: based on absorber area ƞo - 0.832 a1 kJ/hr-m2-K 4.104 a2 kJ/hr-m2-K2 0.0518 Fluid specific heat: Tyfocor kJ/kg-K 3.6–3.9 Fluid density: Tyfocor kg/m3 980–1035 Max. fluid temperature °C 170 Parameter Unit Value Collector type - Direct flow vacuum Number of tubes - 30 Dimensions–gross- mm 1996 × 2127 × 97 Absorber area m2 3.020 Weight–empty- kg 81.4 Fluid content Lt 5.6 Max. operating pressure bar 8 Flow rate -nominal kg/h-m2 80 Efficiency: based on absorber area ƞo - 0.832 a1 kJ/hr-m2-K 4.104 a2 kJ/hr-m2-K2 0.0518 Fluid specific heat: Tyfocor kJ/kg-K 3.6–3.9 Fluid density: Tyfocor kg/m3 980–1035 Max. fluid temperature °C 170 Source: [35] Open in new tab For the chiller, a semi-empirical model that corresponds to the non-standard Type 177, formerly called Type 107, was selected. This model was developed under the method of characteristic equations ΔΔt in TRNSYS 15, within the research project Task 25 of the IEA [33], and has been worked to be compatible with TRNSYS 16. The theory behind this model corresponds to the method of characteristic equations ΔΔt studied by Ziegler et al. [36], which can represent the operation of the absorption cooling cycle through ‘simple algebraic equations’, depending upon the external temperatures of the working fluids [33]. An improvement of the method was developed by Kühn and Ziegler in 2005, which includes a numerical fit of experimental data in order to avoid significant errors in particular at high driving temperatures [33]. As the prototype included the air-cooled version of the Rotartica absorption chiller, the model, however, included the water-cooled version. Rotartica had two models of the same absorption chiller, water cooled –model 0.45, and air-cooled –model 0.45v; therefore the model was set up with the experimental results obtained by Labus [37] that correspond to the water-cooled version of the chiller. Eighteen characteristic parameters of the chiller are needed to simulate its behavior, which was calculated by means of the multiple linear regressions algorithm to support the available experimental data [37,38]. An open cooling tower with counterflow configuration was included using TRNSYS Type 51 subroutine. The thermal performance of the cooling tower is modeled considering the states of the moist air through this equipment and the following empirical correlation [34]: $$\begin{equation}NTU=c{\left(\frac{{\dot{m}}_w}{{\dot{m}}_a}\right)}^{1+n}.\end{equation}$$ (2) The constants c and n correspond to the particular cooling tower configuration used in this model and are 1.684 and −0.391, respectively [39]. For the present model, the cold storage was an insulated water tank, whose performance was modeled using TRNSYS Type 4 routine (represents a sensible thermal storage with vertical stratification effects). The dimensions of the tanks as well as other thermal properties required by the model are shown in Table 3. Table 2 Model parameters of the LiBr/H2O absorption chiller Rotartica 045. Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval—generator kJ/h 27 130 2 rG,I First-order parameter for axis interval—generator kJ/h-K −863.23 3 sG,0 Zero-order parameter for slope calculation—generator kJ/h-K 15.547 4 sG,I First-order parameter for slope calculation -generator kJ/h-K2 34.338 5 rE,0 Zero-order parameter for axis interval—evaporator kJ/h 44 805 6 rE,I First-order parameter for axis interval—evaporator kJ/h-K −1740.7 7 sE,0 Zero-order parameter for slope calculation—evaporator kJ/h-K −413.35 8 sE,I First-order parameter for slope calculation—evaporator kJ/h-K2 41.695 9 B Dühring parameter - 1.2 10 tcw,1 Minimum cooling water inlet temperature °C 25 11 tcw,2 Maximum cooling water inlet temperature °C 40 12 thw,1 Minimum hot water inlet temperature °C 80 13 thw,2 Maximum hot water inlet temperature °C 100 14 tchw,1 Minimum chilled water inlet temperature °C 7 15 tchw,2 Maximum chilled water inlet temperature °C 15 16 mG Rated mass flow rate hot water kg/h 1200 17 mE Rated mass flow rate chilled water kg/h 2000 18 mAC Rated mass flow rate cooling water kg/h 1200 Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval—generator kJ/h 27 130 2 rG,I First-order parameter for axis interval—generator kJ/h-K −863.23 3 sG,0 Zero-order parameter for slope calculation—generator kJ/h-K 15.547 4 sG,I First-order parameter for slope calculation -generator kJ/h-K2 34.338 5 rE,0 Zero-order parameter for axis interval—evaporator kJ/h 44 805 6 rE,I First-order parameter for axis interval—evaporator kJ/h-K −1740.7 7 sE,0 Zero-order parameter for slope calculation—evaporator kJ/h-K −413.35 8 sE,I First-order parameter for slope calculation—evaporator kJ/h-K2 41.695 9 B Dühring parameter - 1.2 10 tcw,1 Minimum cooling water inlet temperature °C 25 11 tcw,2 Maximum cooling water inlet temperature °C 40 12 thw,1 Minimum hot water inlet temperature °C 80 13 thw,2 Maximum hot water inlet temperature °C 100 14 tchw,1 Minimum chilled water inlet temperature °C 7 15 tchw,2 Maximum chilled water inlet temperature °C 15 16 mG Rated mass flow rate hot water kg/h 1200 17 mE Rated mass flow rate chilled water kg/h 2000 18 mAC Rated mass flow rate cooling water kg/h 1200 Source: values were calculated using the methodology described by Albers [38] based on the data gathered by Labus [37]. Open in new tab Table 2 Model parameters of the LiBr/H2O absorption chiller Rotartica 045. Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval—generator kJ/h 27 130 2 rG,I First-order parameter for axis interval—generator kJ/h-K −863.23 3 sG,0 Zero-order parameter for slope calculation—generator kJ/h-K 15.547 4 sG,I First-order parameter for slope calculation -generator kJ/h-K2 34.338 5 rE,0 Zero-order parameter for axis interval—evaporator kJ/h 44 805 6 rE,I First-order parameter for axis interval—evaporator kJ/h-K −1740.7 7 sE,0 Zero-order parameter for slope calculation—evaporator kJ/h-K −413.35 8 sE,I First-order parameter for slope calculation—evaporator kJ/h-K2 41.695 9 B Dühring parameter - 1.2 10 tcw,1 Minimum cooling water inlet temperature °C 25 11 tcw,2 Maximum cooling water inlet temperature °C 40 12 thw,1 Minimum hot water inlet temperature °C 80 13 thw,2 Maximum hot water inlet temperature °C 100 14 tchw,1 Minimum chilled water inlet temperature °C 7 15 tchw,2 Maximum chilled water inlet temperature °C 15 16 mG Rated mass flow rate hot water kg/h 1200 17 mE Rated mass flow rate chilled water kg/h 2000 18 mAC Rated mass flow rate cooling water kg/h 1200 Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval—generator kJ/h 27 130 2 rG,I First-order parameter for axis interval—generator kJ/h-K −863.23 3 sG,0 Zero-order parameter for slope calculation—generator kJ/h-K 15.547 4 sG,I First-order parameter for slope calculation -generator kJ/h-K2 34.338 5 rE,0 Zero-order parameter for axis interval—evaporator kJ/h 44 805 6 rE,I First-order parameter for axis interval—evaporator kJ/h-K −1740.7 7 sE,0 Zero-order parameter for slope calculation—evaporator kJ/h-K −413.35 8 sE,I First-order parameter for slope calculation—evaporator kJ/h-K2 41.695 9 B Dühring parameter - 1.2 10 tcw,1 Minimum cooling water inlet temperature °C 25 11 tcw,2 Maximum cooling water inlet temperature °C 40 12 thw,1 Minimum hot water inlet temperature °C 80 13 thw,2 Maximum hot water inlet temperature °C 100 14 tchw,1 Minimum chilled water inlet temperature °C 7 15 tchw,2 Maximum chilled water inlet temperature °C 15 16 mG Rated mass flow rate hot water kg/h 1200 17 mE Rated mass flow rate chilled water kg/h 2000 18 mAC Rated mass flow rate cooling water kg/h 1200 Source: values were calculated using the methodology described by Albers [38] based on the data gathered by Labus [37]. Open in new tab Table 3 Specifications of the cold-water storage tank. Unit Value Tank volume m3 1 Fluid specific heat kJ/kg-K 4.190 Fluid density kg/m3 1000 Tank loss coefficient kJ/hr-m2-K 3.0 Number of nodes - 5 Height of node 1 m 0.348 Height of node 2 m 0.46 Height of node 3 m 0.505 Height of node 4 m 0.435 Height of node 5 m 0.31 Entering node for hot-side fluid - 1 Entering node for cold-side fluid - 5 Unit Value Tank volume m3 1 Fluid specific heat kJ/kg-K 4.190 Fluid density kg/m3 1000 Tank loss coefficient kJ/hr-m2-K 3.0 Number of nodes - 5 Height of node 1 m 0.348 Height of node 2 m 0.46 Height of node 3 m 0.505 Height of node 4 m 0.435 Height of node 5 m 0.31 Entering node for hot-side fluid - 1 Entering node for cold-side fluid - 5 Open in new tab Table 3 Specifications of the cold-water storage tank. Unit Value Tank volume m3 1 Fluid specific heat kJ/kg-K 4.190 Fluid density kg/m3 1000 Tank loss coefficient kJ/hr-m2-K 3.0 Number of nodes - 5 Height of node 1 m 0.348 Height of node 2 m 0.46 Height of node 3 m 0.505 Height of node 4 m 0.435 Height of node 5 m 0.31 Entering node for hot-side fluid - 1 Entering node for cold-side fluid - 5 Unit Value Tank volume m3 1 Fluid specific heat kJ/kg-K 4.190 Fluid density kg/m3 1000 Tank loss coefficient kJ/hr-m2-K 3.0 Number of nodes - 5 Height of node 1 m 0.348 Height of node 2 m 0.46 Height of node 3 m 0.505 Height of node 4 m 0.435 Height of node 5 m 0.31 Entering node for hot-side fluid - 1 Entering node for cold-side fluid - 5 Open in new tab Regarding the cooling load, the base model was established as an ideal representation to the system installed and tested by Agyenim et al. [31], which was sized to cover an annual cooling demand of 1472 kWh with a peak load of 2.1 kW. The cooling loads registered in the experimental installation corresponded to those of an 82 m3 office at Cardiff University in the summer of 2007. These cooling loads have been approximated and included in the model using the standard component Type 56. The pre-processing program TRNBUILD [34] is used to set up the description of the conditioned space and generate the two files that are used by Type 56 component during the simulations. To obtain the total cooling load for the building, the ‘energy rate’ control is implemented, which is a simplified model of the system that considers a set point of 24°C and unlimited cooling power. Then, an approach where the cooling loads are imposed on the chilled water stream from the cold store has been implemented using equation (3): $$\begin{equation}{T}_{out}={T}_{in}+{Q}_{load}/{\dot{m}}_{chw}{c}_p.\end{equation}$$ (3) Tin and mchw were the temperature and mass flow rate of the chilled water supplied by the cold store tank, Qload is the building cooling load and cp is the specific heat capacity of water. The control strategy of the system was divided into two stages: the cooling generation capacity stage and the cooling distribution stage. Chilled water production depended entirely on the availability of solar radiation, as there was no hot buffer tank or additional heat source to drive the chiller; therefore, to regulate the first stage of the solar-powered absorption chiller, a solar energy rate control is implemented. The differential on/off controller with hysteresis is modeled using TRNSYS Type 2, with a minimum power value of 1000 kJ/hr-m2 and a maximum of 3500 kJ/hr-m2, with a lower dead band insolation difference of 100 kJ/hr-m2. This control, identified as Control-1, commands the operation of Pumps 1–4. The operation of the fan of the cooling tower was regulated with an iterative feedback controller, Control-2, that is modeled with TRNSYS Type 22. Control-2 calculates the control signal to the fan, which determines the airflow through the tower to maintain the temperature of the re-cooling water entering the chiller at the defined set point. The operation of the cooling distribution stage, which consisted of the cold store, the conditioned space and the pump and fan coil unit, depended on whether there was cooling energy demand in the building or not. Therefore, an on/off controller, a thermostat, with a set-point temperature of 24°C had been implemented. 1.4 Validation To assess the accuracy of the proposed model, an empirical validation approach has been utilized. This validation is important because although the models used corresponded to the TRNSYS library, for the chiller component a new model was utilized, calibrated with experimental data. The method involves the comparison of the calculated results from the model developed in TRNSYS to the monitored results from the real installation that operated during the summer of 2007 in Cardiff. The performance of the system was determined by the performance of the absorption chiller, which in turn depended on the three water streams in and out of the equipment; therefore, the inlet parameters used in the model were approximated to the values documented experimentally, including but not limited to the mass flow rates, fluid temperatures and building cooling loads. The performance criteria defined in Table 4 were used to evaluate the results of the simulations. The same performance indicators were used to technically assess the feasibility of the proposed system for the three Ecuadorian cities. Table 4 Performance indicators of the system. Indicator Definition Equation Net collector efficiency Useful solar heat delivered by the solar collector array divided by the total solar radiation on the tilted collector area |${\eta}_{coll}=\frac{Q_{coll}}{I_{sol}}$| Eq. (4) Thermal COP Ratio of the cooling output of the chiller by the required driving heat input |$COP=\frac{Q_e}{Q_{coll}}$| Eq. (5) Solar COP Overall COP of the system: cooling power to the room per available solar power |${COP}_s=\frac{Q_{room}}{Q_{coll}}$| Eq. (6) Indicator Definition Equation Net collector efficiency Useful solar heat delivered by the solar collector array divided by the total solar radiation on the tilted collector area |${\eta}_{coll}=\frac{Q_{coll}}{I_{sol}}$| Eq. (4) Thermal COP Ratio of the cooling output of the chiller by the required driving heat input |$COP=\frac{Q_e}{Q_{coll}}$| Eq. (5) Solar COP Overall COP of the system: cooling power to the room per available solar power |${COP}_s=\frac{Q_{room}}{Q_{coll}}$| Eq. (6) Open in new tab Table 4 Performance indicators of the system. Indicator Definition Equation Net collector efficiency Useful solar heat delivered by the solar collector array divided by the total solar radiation on the tilted collector area |${\eta}_{coll}=\frac{Q_{coll}}{I_{sol}}$| Eq. (4) Thermal COP Ratio of the cooling output of the chiller by the required driving heat input |$COP=\frac{Q_e}{Q_{coll}}$| Eq. (5) Solar COP Overall COP of the system: cooling power to the room per available solar power |${COP}_s=\frac{Q_{room}}{Q_{coll}}$| Eq. (6) Indicator Definition Equation Net collector efficiency Useful solar heat delivered by the solar collector array divided by the total solar radiation on the tilted collector area |${\eta}_{coll}=\frac{Q_{coll}}{I_{sol}}$| Eq. (4) Thermal COP Ratio of the cooling output of the chiller by the required driving heat input |$COP=\frac{Q_e}{Q_{coll}}$| Eq. (5) Solar COP Overall COP of the system: cooling power to the room per available solar power |${COP}_s=\frac{Q_{room}}{Q_{coll}}$| Eq. (6) Open in new tab 1.5 Setup of the model to subtropical climate location The meteorological conditions of Guayaquil, Manta and Puerto Baquerizo Moreno were employed in simulating for both the building cooling demands and cooling power output of the proposed system. The weather data files corresponded to a typical meteorological year Type 2 file, EC-Guayaquil-Aer-842030.tm2, EC-Manta-Eloy-Alfaro-841170.tm2 and C-San-Cristobal-840080.tm2, for Guayaquil, Manta and Puerto Baquerizo Moreno, respectively. For Ecuadorian cities there is no solar radiation data available; therefore the weather dataset only provides interpolated values of solar radiation, based on the geographic position and solar radiation data recorded on the nearest weather stations [40]. The criterion for resizing of the components was that the cooling system must be able to provide the required cooling power to the conditioned space to maintain the room temperature around 24°C, over 80% of the time during working hours (between 08 00 and 20 00). Given that the conditioned space, as well as the cooling loads for the three Ecuadorian cities, is larger, indeed in Guayaquil the peak load was 2-fold of that registered in Cardiff, the total collector area was increased to 24 m2. In the same way as the base model, a position-fixed array of ETCs has been utilized. To maximize the amount of solar heat power generated by the collectors, an azimuth angle of 0° and a tilt angle appropriate to the latitude of the cities were selected; 2°12′S in Guayaquil, 0°57′N in Manta and 0°54′S in Puerto Baquerizo Moreno. A different chiller has been used, due to the increase in cooling demand expected in these cities. The technical data shown in Table 5 correspond to the absorption chiller WEGRACAL SE 15, from the EAW company, using a LiBr/H2O solution as a working pair, and have a nominal capacity of 15 kW [41]. Table 5 Technical data of a 15 kW a +bsorption chiller. Parameter Unit Value Cooling capacity kW 15 Coefficient of performance COP - 0.71 Chilled water Inlet temperature °C 17 Outlet temperature °C 11 Flow rate m3/h 1.9 Heating water Thermal output kW 21 Inlet temperature °C 90 Outlet temperature °C 80 Flow rate m3/h 1.8 Re-cooling water Re-cooling capacity kW 35 Inlet temperature °C 30 Outlet temperature °C 36 Flow rate m3/h 5 Parameter Unit Value Cooling capacity kW 15 Coefficient of performance COP - 0.71 Chilled water Inlet temperature °C 17 Outlet temperature °C 11 Flow rate m3/h 1.9 Heating water Thermal output kW 21 Inlet temperature °C 90 Outlet temperature °C 80 Flow rate m3/h 1.8 Re-cooling water Re-cooling capacity kW 35 Inlet temperature °C 30 Outlet temperature °C 36 Flow rate m3/h 5 Source: [41] Open in new tab Table 5 Technical data of a 15 kW a +bsorption chiller. Parameter Unit Value Cooling capacity kW 15 Coefficient of performance COP - 0.71 Chilled water Inlet temperature °C 17 Outlet temperature °C 11 Flow rate m3/h 1.9 Heating water Thermal output kW 21 Inlet temperature °C 90 Outlet temperature °C 80 Flow rate m3/h 1.8 Re-cooling water Re-cooling capacity kW 35 Inlet temperature °C 30 Outlet temperature °C 36 Flow rate m3/h 5 Parameter Unit Value Cooling capacity kW 15 Coefficient of performance COP - 0.71 Chilled water Inlet temperature °C 17 Outlet temperature °C 11 Flow rate m3/h 1.9 Heating water Thermal output kW 21 Inlet temperature °C 90 Outlet temperature °C 80 Flow rate m3/h 1.8 Re-cooling water Re-cooling capacity kW 35 Inlet temperature °C 30 Outlet temperature °C 36 Flow rate m3/h 5 Source: [41] Open in new tab This chiller has been modeled using the thermal characteristic described in Table 6 that were provided by Albers [38] within the Type 177 routine. Table 6 Model parameters employed in the LiBr/H2O absorption. Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval–generator kJ/h −15 654 2 rG,I First-order parameter for axis interval–generator kJ/h-K 978 3 sG,0 Zero-order parameter for slope calculation–generator kJ/h-K 3203 4 sG,I First-order parameter for slope calculation–generator kJ/h-K2 −60 5 rE,0 Zero-order parameter for axis interval–evaporator kJ/h −1855 6 rE,I First-order parameter for axis interval–evaporator kJ/h-K 55 7 sE,0 Zero-order parameter for slope calculation–evaporator kJ/h-K 2384 8 sE,I First-order parameter for slope calculation–evaporator kJ/h-K2 −41 9 B Dühring parameter - 1.20 10 tcw,1 Minimum cooling water inlet temperature °C 26 11 tcw,2 Maximum cooling water inlet temperature °C 37 12 thw,1 Minimum hot water inlet temperature °C 59 13 thw,2 Maximum hot water inlet temperature °C 96 14 tchw,1 Minimum chilled water inlet temperature °C 5 15 tchw,2 Maximum chilled water inlet temperature °C 25 16 mG Rated mass flow rate hot water kg/h 1940 17 mE Rated mass flow rate chilled water kg/h 1999 18 mAC Rated mass flow rate cooling water kg/h 4975 Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval–generator kJ/h −15 654 2 rG,I First-order parameter for axis interval–generator kJ/h-K 978 3 sG,0 Zero-order parameter for slope calculation–generator kJ/h-K 3203 4 sG,I First-order parameter for slope calculation–generator kJ/h-K2 −60 5 rE,0 Zero-order parameter for axis interval–evaporator kJ/h −1855 6 rE,I First-order parameter for axis interval–evaporator kJ/h-K 55 7 sE,0 Zero-order parameter for slope calculation–evaporator kJ/h-K 2384 8 sE,I First-order parameter for slope calculation–evaporator kJ/h-K2 −41 9 B Dühring parameter - 1.20 10 tcw,1 Minimum cooling water inlet temperature °C 26 11 tcw,2 Maximum cooling water inlet temperature °C 37 12 thw,1 Minimum hot water inlet temperature °C 59 13 thw,2 Maximum hot water inlet temperature °C 96 14 tchw,1 Minimum chilled water inlet temperature °C 5 15 tchw,2 Maximum chilled water inlet temperature °C 25 16 mG Rated mass flow rate hot water kg/h 1940 17 mE Rated mass flow rate chilled water kg/h 1999 18 mAC Rated mass flow rate cooling water kg/h 4975 Source: [38] Open in new tab Table 6 Model parameters employed in the LiBr/H2O absorption. Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval–generator kJ/h −15 654 2 rG,I First-order parameter for axis interval–generator kJ/h-K 978 3 sG,0 Zero-order parameter for slope calculation–generator kJ/h-K 3203 4 sG,I First-order parameter for slope calculation–generator kJ/h-K2 −60 5 rE,0 Zero-order parameter for axis interval–evaporator kJ/h −1855 6 rE,I First-order parameter for axis interval–evaporator kJ/h-K 55 7 sE,0 Zero-order parameter for slope calculation–evaporator kJ/h-K 2384 8 sE,I First-order parameter for slope calculation–evaporator kJ/h-K2 −41 9 B Dühring parameter - 1.20 10 tcw,1 Minimum cooling water inlet temperature °C 26 11 tcw,2 Maximum cooling water inlet temperature °C 37 12 thw,1 Minimum hot water inlet temperature °C 59 13 thw,2 Maximum hot water inlet temperature °C 96 14 tchw,1 Minimum chilled water inlet temperature °C 5 15 tchw,2 Maximum chilled water inlet temperature °C 25 16 mG Rated mass flow rate hot water kg/h 1940 17 mE Rated mass flow rate chilled water kg/h 1999 18 mAC Rated mass flow rate cooling water kg/h 4975 Id. Name Description Unit Value 1 rG,0 Zero-order parameter for axis interval–generator kJ/h −15 654 2 rG,I First-order parameter for axis interval–generator kJ/h-K 978 3 sG,0 Zero-order parameter for slope calculation–generator kJ/h-K 3203 4 sG,I First-order parameter for slope calculation–generator kJ/h-K2 −60 5 rE,0 Zero-order parameter for axis interval–evaporator kJ/h −1855 6 rE,I First-order parameter for axis interval–evaporator kJ/h-K 55 7 sE,0 Zero-order parameter for slope calculation–evaporator kJ/h-K 2384 8 sE,I First-order parameter for slope calculation–evaporator kJ/h-K2 −41 9 B Dühring parameter - 1.20 10 tcw,1 Minimum cooling water inlet temperature °C 26 11 tcw,2 Maximum cooling water inlet temperature °C 37 12 thw,1 Minimum hot water inlet temperature °C 59 13 thw,2 Maximum hot water inlet temperature °C 96 14 tchw,1 Minimum chilled water inlet temperature °C 5 15 tchw,2 Maximum chilled water inlet temperature °C 25 16 mG Rated mass flow rate hot water kg/h 1940 17 mE Rated mass flow rate chilled water kg/h 1999 18 mAC Rated mass flow rate cooling water kg/h 4975 Source: [38] Open in new tab In the same fashion as the base model, a cooling tower and a cold store tank were included in the set-up for the Ecuadorian cities. However, their heat rejection and store capacities, respectively, have been increased. This has been accomplished increasing the flow rates of the re-cooling water stream and air in the cooling tower, and the size of the tank, with a volume of 2 m3. The cooling load is included in the system using the standard component Type 56. The load was modeled as a single-story office building of 450 m3 without internal partitions and with external structures as described in Table 7. The construction elements are of similar characteristics of those typically used in local constructions in Ecuador. An estimated infiltration rate of 0.25 air change rate per hour (ACH) has been included in the building model. The internal gains arise from three sources: people, office equipment and artificial lighting. To evaluate the variations of the performance of the proposed system for the three cities, the internal loads remained the same. Table 7 Construction materials and U-values. Wall type Layers U-value (W/m2-K) Ground Floor 0.005 m 0.834 Stone 0.060 m Silence 0.040 m Concrete 0.240 m External wall Plaster 0.015 m 2.554 Concrete hollow blocks 0.102 m Plaster 0.015 m Roof and ceiling Light concrete element 0.050 m 0.958 Air layer Expanded polystyrene 0.010 m Plasterboard 0.020 m Windows: 25% of the area on North and South walls Single glazed framed window 5.68 Wall type Layers U-value (W/m2-K) Ground Floor 0.005 m 0.834 Stone 0.060 m Silence 0.040 m Concrete 0.240 m External wall Plaster 0.015 m 2.554 Concrete hollow blocks 0.102 m Plaster 0.015 m Roof and ceiling Light concrete element 0.050 m 0.958 Air layer Expanded polystyrene 0.010 m Plasterboard 0.020 m Windows: 25% of the area on North and South walls Single glazed framed window 5.68 Open in new tab Table 7 Construction materials and U-values. Wall type Layers U-value (W/m2-K) Ground Floor 0.005 m 0.834 Stone 0.060 m Silence 0.040 m Concrete 0.240 m External wall Plaster 0.015 m 2.554 Concrete hollow blocks 0.102 m Plaster 0.015 m Roof and ceiling Light concrete element 0.050 m 0.958 Air layer Expanded polystyrene 0.010 m Plasterboard 0.020 m Windows: 25% of the area on North and South walls Single glazed framed window 5.68 Wall type Layers U-value (W/m2-K) Ground Floor 0.005 m 0.834 Stone 0.060 m Silence 0.040 m Concrete 0.240 m External wall Plaster 0.015 m 2.554 Concrete hollow blocks 0.102 m Plaster 0.015 m Roof and ceiling Light concrete element 0.050 m 0.958 Air layer Expanded polystyrene 0.010 m Plasterboard 0.020 m Windows: 25% of the area on North and South walls Single glazed framed window 5.68 Open in new tab The first stage control, for the cooling generation capacity, was implemented in the same fashion as in the base model, using a differential controller TRNSYS Type 2 for Pumps 1–4, and a feedback controller TRNSYS Type 22 to regulate the air flow in the cooling tower. The minimum level of solar radiation to drive the pumps of the solar subsystem was 500 kJ/hr.m2. Regarding the cooling distribution stage, similarly to the base model, a thermostat is implemented. The thermostat was set for a temperature range between 22°C and 24°C, which reduces system intermittence. To couple the model of the building with the solar-powered absorption system, additional elements were included: a single speed fan, Type 112b; a simplified model for the cooling coil, Type 32; and two components to calculate the psychometrics properties of the moist air, which are modeled with TRNSYS Type 33. This approach, known as ‘temperature level’ control, was selected for this model to simulate with more detail the cooling equipment. Additional control of the room thermostat was implemented. This control was scheduled so that the system operates only in office hours (between 08 00 and 20 00). The diagram flow of the model adapted to the new location is shown in Figure 6. Figure 6 Open in new tabDownload slide Flow diagram of the solar-powered absorption cooling system in TRNSYS. Figure 6 Open in new tabDownload slide Flow diagram of the solar-powered absorption cooling system in TRNSYS. 2 RESULTS AND ANALYSIS Two models of a solar-powered absorption cooling system were developed for the locations: Cardiff in the UK and the second for Guayaquil, Manta and Puerto Baquerizo Moreno in Ecuador. A short time-step of 5 minutes was used in the simulations. As it was mentioned the meteorology for the based model was taken from TRNSYS database; therefore a minor deviation from the experimental results is expected. The selection of the day for the comparison and validation of the model with the experimental data at Cardiff, the UK on 24 August 2007, was based on the global solar radiation profile. The selected day was 30 August. The difference between insolation profile of the theoretical (30 August) and experimental (24 August) meteorological data between 10 00 and 16 00 is less than 10%. To validate the model the following equation was used to calculate the relative error: $$\begin{equation}\mathrm{error}\%=\frac{\left({\mathrm{value}}_{\mathrm{sim}}-{\mathrm{value}}_{\mathrm{exp}}\right)}{{\mathrm{value}}_{\mathrm{exp}}}\times 100,\end{equation}$$ (7) where valuesim and valueexp are the performance parameters calculated within the model and values registered by Agyenim et al. [31] during the experimental operation, respectively. The performance of the entire system is strongly influenced by the three fluid streams entering the absorption chiller. This implies that to compare the results of the model with the measured values, inlet parameters to the model, such as the total solar radiation on the collector surface and the building cooling loads, are required to be equal, or else, very similar to the values documented during the experimental work. Hence, the conditioned space model was developed so that the cooling demand daily profile follows as close as possible the actual profile (Figure 7). Figure 7 Open in new tabDownload slide Measure and simulated total solar radiation power on the collector surface and cooling load for an 82 m3 office. Figure 7 Open in new tabDownload slide Measure and simulated total solar radiation power on the collector surface and cooling load for an 82 m3 office. The deviation between simulated and experimental data for the solar input power and the cooling power output of the entire system is small, particularly between 10 00 and 16 00 (Figure 7). However, the profiles differ significantly in the early morning and late afternoon, when the availability of solar radiation is lower (Tables 8 and 9). Table 8 Comparison of operational temperatures of the system. System components Between 10 00 and 16 00 All day Average °C Avg. error Min. °C Max. °C Average °C Avg. error Sim. Exp. Sim. Exp. Sim. Exp. Sim. Exp. Collector outlet 84.7 81.7 4% 68.7 63.3 91.5 92.0 68.2 65.5 20% Chiller generator inlet 82.2 78.0 5% 66.7 60.1 88.8 88.1 65.0 63.9 30% Chiller evaporator outlet 10.9 9.8 12% 9.4 7.3 14.4 14.5 14.6 13.1 14% Fan coil inlet 11.0 10.5 6% 9.4 8.0 14.7 15.2 14.1 12.6 13% System components Between 10 00 and 16 00 All day Average °C Avg. error Min. °C Max. °C Average °C Avg. error Sim. Exp. Sim. Exp. Sim. Exp. Sim. Exp. Collector outlet 84.7 81.7 4% 68.7 63.3 91.5 92.0 68.2 65.5 20% Chiller generator inlet 82.2 78.0 5% 66.7 60.1 88.8 88.1 65.0 63.9 30% Chiller evaporator outlet 10.9 9.8 12% 9.4 7.3 14.4 14.5 14.6 13.1 14% Fan coil inlet 11.0 10.5 6% 9.4 8.0 14.7 15.2 14.1 12.6 13% Open in new tab Table 8 Comparison of operational temperatures of the system. System components Between 10 00 and 16 00 All day Average °C Avg. error Min. °C Max. °C Average °C Avg. error Sim. Exp. Sim. Exp. Sim. Exp. Sim. Exp. Collector outlet 84.7 81.7 4% 68.7 63.3 91.5 92.0 68.2 65.5 20% Chiller generator inlet 82.2 78.0 5% 66.7 60.1 88.8 88.1 65.0 63.9 30% Chiller evaporator outlet 10.9 9.8 12% 9.4 7.3 14.4 14.5 14.6 13.1 14% Fan coil inlet 11.0 10.5 6% 9.4 8.0 14.7 15.2 14.1 12.6 13% System components Between 10 00 and 16 00 All day Average °C Avg. error Min. °C Max. °C Average °C Avg. error Sim. Exp. Sim. Exp. Sim. Exp. Sim. Exp. Collector outlet 84.7 81.7 4% 68.7 63.3 91.5 92.0 68.2 65.5 20% Chiller generator inlet 82.2 78.0 5% 66.7 60.1 88.8 88.1 65.0 63.9 30% Chiller evaporator outlet 10.9 9.8 12% 9.4 7.3 14.4 14.5 14.6 13.1 14% Fan coil inlet 11.0 10.5 6% 9.4 8.0 14.7 15.2 14.1 12.6 13% Open in new tab Table 9 Comparison of performance parameters of the system. Parameters Between 10 00 and 16 00 All day Sim. Exp. Avg. error SD Sim. Exp. Avg. error SD Sim. Exp. Sim. Exp. Collector net efficiency 0.68 0.66 0.2% 1% 10% 0.67 0.59 5% 2% 27% Chiller thermal COP 0.73 0.64 24% 6% 17% 0.84 0.18 51% 22% 336% System solar COP 0.34 0.37 2% 7% 17% 0.51 0.99 14% 160% 204% Parameters Between 10 00 and 16 00 All day Sim. Exp. Avg. error SD Sim. Exp. Avg. error SD Sim. Exp. Sim. Exp. Collector net efficiency 0.68 0.66 0.2% 1% 10% 0.67 0.59 5% 2% 27% Chiller thermal COP 0.73 0.64 24% 6% 17% 0.84 0.18 51% 22% 336% System solar COP 0.34 0.37 2% 7% 17% 0.51 0.99 14% 160% 204% Open in new tab Table 9 Comparison of performance parameters of the system. Parameters Between 10 00 and 16 00 All day Sim. Exp. Avg. error SD Sim. Exp. Avg. error SD Sim. Exp. Sim. Exp. Collector net efficiency 0.68 0.66 0.2% 1% 10% 0.67 0.59 5% 2% 27% Chiller thermal COP 0.73 0.64 24% 6% 17% 0.84 0.18 51% 22% 336% System solar COP 0.34 0.37 2% 7% 17% 0.51 0.99 14% 160% 204% Parameters Between 10 00 and 16 00 All day Sim. Exp. Avg. error SD Sim. Exp. Avg. error SD Sim. Exp. Sim. Exp. Collector net efficiency 0.68 0.66 0.2% 1% 10% 0.67 0.59 5% 2% 27% Chiller thermal COP 0.73 0.64 24% 6% 17% 0.84 0.18 51% 22% 336% System solar COP 0.34 0.37 2% 7% 17% 0.51 0.99 14% 160% 204% Open in new tab Figure 8 shows the useful thermal power output of the collector field and its efficiency. The collector output follows closely the total solar insolation; the efficiency remains around 0.66–0.68 between the hours with higher solar radiation, from 10 00 to 16 00. Figures 9 and 10 show the comparison between the operational temperatures of the absorption chiller, the cooling capacity and its thermal COP (coefficient of performance). The simulated results predict the experimental data at an acceptable error of approximately 12%. On the other hand, the generator temperature is to some extent less than expected, considering that the solar radiation power in the simulation results is noticeably higher than measured values. Despite the above-mentioned, the thermal performance of the chiller was not reduced, in fact, is slightly better than its experimental counterpart (Figure 10). This could be attributed to the use of a water-cooled version of the absorption chiller in the model, instead of the air-cooled chiller that was used in the installation at Cardiff University. This observation is similar to what has been concluded in previous works [12], where they also acknowledge that the influence of the heat rejection temperature determines not only the thermal performance of the absorption chiller but also its electricity consumption, in particular in part-load operation. The model of the chiller used in this work, Type 177, does not include the calculation of its electricity consumption. Electricity consumption was therefore not simulated. It was, however, calculated from technical data supplied by manufacturers. It was estimated that the electricity consumption rate of the model was ~1.23 kW, which agrees with the results presented by Agyenim et al. [31]. The average error increases if it is calculated for the entire time range, from 06 00 to 20 00. This is a consequence of the unsteady operation of the system in the early hours of the morning and late afternoon, which is more accentuated in the real system in comparison with the theoretical model, where inertial effects in the heat exchanger and absorption chiller were not included. The collector efficiency shows good agreement between the simulated and measured values, even for the entire day, likewise, the solar COP of the system (Figure 11). The simulated thermal COP of the chiller, however, differs almost 25% from the experimental data, which may be due to the omission of some inertial effects as discussed before. Figure 8 Open in new tabDownload slide Solar thermal collection system: power output and net efficiency. Figure 8 Open in new tabDownload slide Solar thermal collection system: power output and net efficiency. Figure 9 Open in new tabDownload slide Measured and simulated inlet and outlet fluid temperatures of the absorption chiller. Figure 9 Open in new tabDownload slide Measured and simulated inlet and outlet fluid temperatures of the absorption chiller. Figure 10 Open in new tabDownload slide Measured and simulated absorption chiller cooling capacity and thermal COP. Figure 10 Open in new tabDownload slide Measured and simulated absorption chiller cooling capacity and thermal COP. Figure 11 Open in new tabDownload slide Solar COP of the system. Figure 11 Open in new tabDownload slide Solar COP of the system. 2.1 Performance of the solar-powered cooling system in Ecuadorian cities Figures 12–14 show the performance parameters on a typical hot day in Guayaquil. Over a 24-hour period, the performance of the system follows the same trend as the base model at Cardiff. However, as the solar resource is higher, the collector and chiller output achieves higher and lower temperatures, respectively, than those of the base model at Cardiff. The efficiency of the collector array registered similar values for both cities; the thermal COP of the chiller is better for Guayaquil. Figure 12 Open in new tabDownload slide Hourly variation of solar radiation, ambient temperature, collector output, chiller output, tank average and room temperature on a typical hot day (30 March) with 976 W/m2 peak radiation in Guayaquil, Ecuador. Figure 12 Open in new tabDownload slide Hourly variation of solar radiation, ambient temperature, collector output, chiller output, tank average and room temperature on a typical hot day (30 March) with 976 W/m2 peak radiation in Guayaquil, Ecuador. Figure 13 Open in new tabDownload slide Hourly variation of solar power available, collector power output and collector efficiency, on a typical hot day (19 March) in Guayaquil, Ecuador. Figure 13 Open in new tabDownload slide Hourly variation of solar power available, collector power output and collector efficiency, on a typical hot day (19 March) in Guayaquil, Ecuador. Figure 14 Open in new tabDownload slide Hourly variation of chiller cooling power output and Thermal COP, on a typical hot day with 976 W/m2 peak radiation in Guayaquil, Ecuador. Figure 14 Open in new tabDownload slide Hourly variation of chiller cooling power output and Thermal COP, on a typical hot day with 976 W/m2 peak radiation in Guayaquil, Ecuador. Over a 24-hour period, there are 11 hours of solar resource. The temperature of the water inside the collector starts to increase after 07 00, and chilled water production starts from 08 00. The minimum hot water temperature for the chiller to start to produce cooling is 59°C; the level of solar radiation prior to achieving this temperature is 98 W/m2. To drive the chiller a minimum power of solar radiation required is 155 W/m2. The maximum chiller power output of 10.15 kW is reached at 12 25. The average efficiency of the collector for the day is 0.59, and during the period between 10 s00 and 16 00, when the solar radiation is less variable, the average efficiency is 0.65. The average thermal COP of the chiller over a period of 24 hours is 0.60, and between 10 00 and 16 00 is 0.63, which is close to the nominal value of 0.71 for the selected chiller. The solar COP is an efficiency parameter comprising the entire system; which relates the cooling power to the room with the collector power output. The average solar COP is 0.45 between 10 00 and 16 00, and 0.74 over a 24-hour period. The value increased on the last hours of the afternoon due to the availability of stored energy in the form of chilled water in the tank. The daily performance of the system for the entire year in Guayaquil and in the other two cities, Manta and San Cristobal, show the same trend, particularly on the warmest months, where the solar resource and cooling demand are larger. To compare the performance of the system in the three Ecuadorian cities, annual simulations were performed. As it was expected, the warmest months, between January and May, are as well the months with higher loads. In Guayaquil and Manta, March is the warmest month, and in Puerto Baquerizo (San Cristobal), April. A summary of the results on the warmest months is shown in Table 10. Table 10 Operational parameters of the Solar Cooling System in Guayaquil, Manta and San Cristobal (between 8 00 and 20 00). Parameter Unit Average Maximum Minimum Deviation Guayaquil/March  Ambient temperature °C 28.65 33.00 22.64 2.05  Wet bulb temperature °C 24.24 26.88 20.57 1.19  Room temperature °C 22.61 26.85 19.52 1.00  Cooling power to room kW 4.36 13.74 0.00 4.45  Total Insolation on collector W/m2 488.43 1043.88 0.00 330.01  Collector outlet temperature °C 96.25 132.39 42.97 20.42  Collector power output kW 9.22 16.73 0.86 4.28  Efficiency of the collector - 0.61 0.73 0.31 0.06  Chiller inlet generator temperature °C 80.75 107.10 37.64 13.50  Chiller inlet re-cooling temperature °C 34.88 37.51 20.49 1.51  Chiller outlet evaporator temperature °C 12.05 22.64 5.09 4.15  Generator inlet power kW 9.21 16.55 0.85 4.33  Chiller cooling capacity kW 5.89 11.95 0.00 2.89  Chiller heat rejection kW 15.39 28.23 0.96 6.99  Chiller thermal COP - 0.60 1.25 0.00 0.12  Average tank temperature °C 13.33 23.59 5.72 4.26  Solar COP - 0.71 11.21 0.00 1.19 Manta/March  Ambient temperature °C 27.40 31.10 21.53 1.62  Wet bulb temperature °C 23.98 26.39 19.89 1.20  Room temperature °C 23.36 26.65 18.96 1.18  Cooling power to room kW 3.75 11.94 0.00 3.02  Total Insolation on collector W/m2 493.52 1045.07 0.00 334.07  Collector outlet temperature °C 103.03 137.31 63.31 20.46  Collector power output kW 8.76 16.42 0.53 4.25  Efficiency of the collector - 0.56 0.73 0.20 0.08  Chiller inlet generator temperature °C 88.31 112.70 60.88 13.47  Chiller inlet re-cooling temperature °C 40.94 43.62 21.93 1.31  Chiller outlet evaporator temperature °C 15.67 25.74 7.01 4.09  Generator inlet power kW 8.68 16.20 0.43 4.28  Chiller cooling capacity kW 5.24 12.82 0.03 2.74  Chiller heat rejection kW 13.80 29.02 0.01 7.09  Chiller thermal COP - 0.53 0.79 0.01 0.12  Average tank temperature °C 17.02 25.70 7.55 4.11  Solar COP - 0.69 16.20 0.00 1.14 San Cristobal/April  Ambient temperature °C 28.97 33.70 23.17 1.91  Wet bulb temperature °C 21.92 24.91 19.03 1.13  Room temperature °C 22.61 27.85 19.85 1.12  Cooling power to room kW 4.46 12.67 0.00 4.52  Total insolation on collector W/m2 634.39 1097.28 0.00 370.99  Collector outlet temperature °C 118.00 149.56 65.79 24.35  Collector power output kW 10.60 16.03 0.40 4.66  Efficiency of the collector - 0.57 0.66 0.15 0.08  Chiller inlet generator temperature °C 100.19 124.33 62.82 16.68  Chiller inlet re-cooling temperature °C 41.63 45.00 37.31 1.03  Chiller outlet evaporator temperature °C 11.57 22.62 5.15 3.72  Generator inlet power kW 10.59 16.05 0.36 4.71  Chiller cooling capacity kW 6.25 10.09 0.00 2.97  Chiller heat rejection kW 16.53 26.07 0.02 7.90  Chiller thermal COP - 0.50 0.65 0.00 0.14  Average tank temperature °C 12.53 22.34 5.73 3.68  Solar COP - 0.65 20.37 0.00 1.34 Parameter Unit Average Maximum Minimum Deviation Guayaquil/March  Ambient temperature °C 28.65 33.00 22.64 2.05  Wet bulb temperature °C 24.24 26.88 20.57 1.19  Room temperature °C 22.61 26.85 19.52 1.00  Cooling power to room kW 4.36 13.74 0.00 4.45  Total Insolation on collector W/m2 488.43 1043.88 0.00 330.01  Collector outlet temperature °C 96.25 132.39 42.97 20.42  Collector power output kW 9.22 16.73 0.86 4.28  Efficiency of the collector - 0.61 0.73 0.31 0.06  Chiller inlet generator temperature °C 80.75 107.10 37.64 13.50  Chiller inlet re-cooling temperature °C 34.88 37.51 20.49 1.51  Chiller outlet evaporator temperature °C 12.05 22.64 5.09 4.15  Generator inlet power kW 9.21 16.55 0.85 4.33  Chiller cooling capacity kW 5.89 11.95 0.00 2.89  Chiller heat rejection kW 15.39 28.23 0.96 6.99  Chiller thermal COP - 0.60 1.25 0.00 0.12  Average tank temperature °C 13.33 23.59 5.72 4.26  Solar COP - 0.71 11.21 0.00 1.19 Manta/March  Ambient temperature °C 27.40 31.10 21.53 1.62  Wet bulb temperature °C 23.98 26.39 19.89 1.20  Room temperature °C 23.36 26.65 18.96 1.18  Cooling power to room kW 3.75 11.94 0.00 3.02  Total Insolation on collector W/m2 493.52 1045.07 0.00 334.07  Collector outlet temperature °C 103.03 137.31 63.31 20.46  Collector power output kW 8.76 16.42 0.53 4.25  Efficiency of the collector - 0.56 0.73 0.20 0.08  Chiller inlet generator temperature °C 88.31 112.70 60.88 13.47  Chiller inlet re-cooling temperature °C 40.94 43.62 21.93 1.31  Chiller outlet evaporator temperature °C 15.67 25.74 7.01 4.09  Generator inlet power kW 8.68 16.20 0.43 4.28  Chiller cooling capacity kW 5.24 12.82 0.03 2.74  Chiller heat rejection kW 13.80 29.02 0.01 7.09  Chiller thermal COP - 0.53 0.79 0.01 0.12  Average tank temperature °C 17.02 25.70 7.55 4.11  Solar COP - 0.69 16.20 0.00 1.14 San Cristobal/April  Ambient temperature °C 28.97 33.70 23.17 1.91  Wet bulb temperature °C 21.92 24.91 19.03 1.13  Room temperature °C 22.61 27.85 19.85 1.12  Cooling power to room kW 4.46 12.67 0.00 4.52  Total insolation on collector W/m2 634.39 1097.28 0.00 370.99  Collector outlet temperature °C 118.00 149.56 65.79 24.35  Collector power output kW 10.60 16.03 0.40 4.66  Efficiency of the collector - 0.57 0.66 0.15 0.08  Chiller inlet generator temperature °C 100.19 124.33 62.82 16.68  Chiller inlet re-cooling temperature °C 41.63 45.00 37.31 1.03  Chiller outlet evaporator temperature °C 11.57 22.62 5.15 3.72  Generator inlet power kW 10.59 16.05 0.36 4.71  Chiller cooling capacity kW 6.25 10.09 0.00 2.97  Chiller heat rejection kW 16.53 26.07 0.02 7.90  Chiller thermal COP - 0.50 0.65 0.00 0.14  Average tank temperature °C 12.53 22.34 5.73 3.68  Solar COP - 0.65 20.37 0.00 1.34 Open in new tab Table 10 Operational parameters of the Solar Cooling System in Guayaquil, Manta and San Cristobal (between 8 00 and 20 00). Parameter Unit Average Maximum Minimum Deviation Guayaquil/March  Ambient temperature °C 28.65 33.00 22.64 2.05  Wet bulb temperature °C 24.24 26.88 20.57 1.19  Room temperature °C 22.61 26.85 19.52 1.00  Cooling power to room kW 4.36 13.74 0.00 4.45  Total Insolation on collector W/m2 488.43 1043.88 0.00 330.01  Collector outlet temperature °C 96.25 132.39 42.97 20.42  Collector power output kW 9.22 16.73 0.86 4.28  Efficiency of the collector - 0.61 0.73 0.31 0.06  Chiller inlet generator temperature °C 80.75 107.10 37.64 13.50  Chiller inlet re-cooling temperature °C 34.88 37.51 20.49 1.51  Chiller outlet evaporator temperature °C 12.05 22.64 5.09 4.15  Generator inlet power kW 9.21 16.55 0.85 4.33  Chiller cooling capacity kW 5.89 11.95 0.00 2.89  Chiller heat rejection kW 15.39 28.23 0.96 6.99  Chiller thermal COP - 0.60 1.25 0.00 0.12  Average tank temperature °C 13.33 23.59 5.72 4.26  Solar COP - 0.71 11.21 0.00 1.19 Manta/March  Ambient temperature °C 27.40 31.10 21.53 1.62  Wet bulb temperature °C 23.98 26.39 19.89 1.20  Room temperature °C 23.36 26.65 18.96 1.18  Cooling power to room kW 3.75 11.94 0.00 3.02  Total Insolation on collector W/m2 493.52 1045.07 0.00 334.07  Collector outlet temperature °C 103.03 137.31 63.31 20.46  Collector power output kW 8.76 16.42 0.53 4.25  Efficiency of the collector - 0.56 0.73 0.20 0.08  Chiller inlet generator temperature °C 88.31 112.70 60.88 13.47  Chiller inlet re-cooling temperature °C 40.94 43.62 21.93 1.31  Chiller outlet evaporator temperature °C 15.67 25.74 7.01 4.09  Generator inlet power kW 8.68 16.20 0.43 4.28  Chiller cooling capacity kW 5.24 12.82 0.03 2.74  Chiller heat rejection kW 13.80 29.02 0.01 7.09  Chiller thermal COP - 0.53 0.79 0.01 0.12  Average tank temperature °C 17.02 25.70 7.55 4.11  Solar COP - 0.69 16.20 0.00 1.14 San Cristobal/April  Ambient temperature °C 28.97 33.70 23.17 1.91  Wet bulb temperature °C 21.92 24.91 19.03 1.13  Room temperature °C 22.61 27.85 19.85 1.12  Cooling power to room kW 4.46 12.67 0.00 4.52  Total insolation on collector W/m2 634.39 1097.28 0.00 370.99  Collector outlet temperature °C 118.00 149.56 65.79 24.35  Collector power output kW 10.60 16.03 0.40 4.66  Efficiency of the collector - 0.57 0.66 0.15 0.08  Chiller inlet generator temperature °C 100.19 124.33 62.82 16.68  Chiller inlet re-cooling temperature °C 41.63 45.00 37.31 1.03  Chiller outlet evaporator temperature °C 11.57 22.62 5.15 3.72  Generator inlet power kW 10.59 16.05 0.36 4.71  Chiller cooling capacity kW 6.25 10.09 0.00 2.97  Chiller heat rejection kW 16.53 26.07 0.02 7.90  Chiller thermal COP - 0.50 0.65 0.00 0.14  Average tank temperature °C 12.53 22.34 5.73 3.68  Solar COP - 0.65 20.37 0.00 1.34 Parameter Unit Average Maximum Minimum Deviation Guayaquil/March  Ambient temperature °C 28.65 33.00 22.64 2.05  Wet bulb temperature °C 24.24 26.88 20.57 1.19  Room temperature °C 22.61 26.85 19.52 1.00  Cooling power to room kW 4.36 13.74 0.00 4.45  Total Insolation on collector W/m2 488.43 1043.88 0.00 330.01  Collector outlet temperature °C 96.25 132.39 42.97 20.42  Collector power output kW 9.22 16.73 0.86 4.28  Efficiency of the collector - 0.61 0.73 0.31 0.06  Chiller inlet generator temperature °C 80.75 107.10 37.64 13.50  Chiller inlet re-cooling temperature °C 34.88 37.51 20.49 1.51  Chiller outlet evaporator temperature °C 12.05 22.64 5.09 4.15  Generator inlet power kW 9.21 16.55 0.85 4.33  Chiller cooling capacity kW 5.89 11.95 0.00 2.89  Chiller heat rejection kW 15.39 28.23 0.96 6.99  Chiller thermal COP - 0.60 1.25 0.00 0.12  Average tank temperature °C 13.33 23.59 5.72 4.26  Solar COP - 0.71 11.21 0.00 1.19 Manta/March  Ambient temperature °C 27.40 31.10 21.53 1.62  Wet bulb temperature °C 23.98 26.39 19.89 1.20  Room temperature °C 23.36 26.65 18.96 1.18  Cooling power to room kW 3.75 11.94 0.00 3.02  Total Insolation on collector W/m2 493.52 1045.07 0.00 334.07  Collector outlet temperature °C 103.03 137.31 63.31 20.46  Collector power output kW 8.76 16.42 0.53 4.25  Efficiency of the collector - 0.56 0.73 0.20 0.08  Chiller inlet generator temperature °C 88.31 112.70 60.88 13.47  Chiller inlet re-cooling temperature °C 40.94 43.62 21.93 1.31  Chiller outlet evaporator temperature °C 15.67 25.74 7.01 4.09  Generator inlet power kW 8.68 16.20 0.43 4.28  Chiller cooling capacity kW 5.24 12.82 0.03 2.74  Chiller heat rejection kW 13.80 29.02 0.01 7.09  Chiller thermal COP - 0.53 0.79 0.01 0.12  Average tank temperature °C 17.02 25.70 7.55 4.11  Solar COP - 0.69 16.20 0.00 1.14 San Cristobal/April  Ambient temperature °C 28.97 33.70 23.17 1.91  Wet bulb temperature °C 21.92 24.91 19.03 1.13  Room temperature °C 22.61 27.85 19.85 1.12  Cooling power to room kW 4.46 12.67 0.00 4.52  Total insolation on collector W/m2 634.39 1097.28 0.00 370.99  Collector outlet temperature °C 118.00 149.56 65.79 24.35  Collector power output kW 10.60 16.03 0.40 4.66  Efficiency of the collector - 0.57 0.66 0.15 0.08  Chiller inlet generator temperature °C 100.19 124.33 62.82 16.68  Chiller inlet re-cooling temperature °C 41.63 45.00 37.31 1.03  Chiller outlet evaporator temperature °C 11.57 22.62 5.15 3.72  Generator inlet power kW 10.59 16.05 0.36 4.71  Chiller cooling capacity kW 6.25 10.09 0.00 2.97  Chiller heat rejection kW 16.53 26.07 0.02 7.90  Chiller thermal COP - 0.50 0.65 0.00 0.14  Average tank temperature °C 12.53 22.34 5.73 3.68  Solar COP - 0.65 20.37 0.00 1.34 Open in new tab The average performance of the system in Guayaquil and Manta are similar between each other. In San Cristobal, the level of insolation is higher, and the relative humidity is lower, in comparison with the other two cities. In San Cristobal, due to the higher solar resource, the amount of energy harvested from the collector array is higher, as well as the chiller cooling capacity and the cooling power to the room. However, the average performance of the system (efficiency of the collector, thermal COP of the chiller and solar COP) is noticeably lower in comparison with Guayaquil and Manta (Table 10). Figures 15–19 show the annual simulation results, comparing the three cities. The total annual energy output from the collector is 26 265 kWh, 26 677 kWh and 26 516 kWh, respectively in Guayaquil, Manta and San Cristobal, which represents correspondingly 58%, 56%, and 53% of the total incident solar energy on the collector array. Figure 15 Open in new tabDownload slide Collector power output in Guayaquil, Manta and San Cristobal. Figure 15 Open in new tabDownload slide Collector power output in Guayaquil, Manta and San Cristobal. Figure 16 Open in new tabDownload slide Cooling load in Guayaquil, Manta and San Cristobal. Figure 16 Open in new tabDownload slide Cooling load in Guayaquil, Manta and San Cristobal. Figure 17 Open in new tabDownload slide Monthly variation of collector array efficiency in Guayaquil, Manta and San Cristobal. Figure 17 Open in new tabDownload slide Monthly variation of collector array efficiency in Guayaquil, Manta and San Cristobal. Figure 18 Open in new tabDownload slide Monthly variation of absorption chiller COP in Guayaquil, Manta and San Cristobal. Figure 18 Open in new tabDownload slide Monthly variation of absorption chiller COP in Guayaquil, Manta and San Cristobal. Figure 19 Open in new tabDownload slide Monthly variation of system solar COP in Guayaquil, Manta and San Cristobal. Figure 19 Open in new tabDownload slide Monthly variation of system solar COP in Guayaquil, Manta and San Cristobal. Table 11 Comparison with literature for absorption cooling systems in tropical climates. Ref. Author Year Location Building type Floor area (m2) Solar collector Collector area (m2) A/C unit Working pair Cooling capacity (kW) Backup system Average thermal COP Comments [13] Abed et al. 2017 Malaysia Single zone building ETC SS with single ejector NH3-H2O 3–5 — 0.127–0.282 SS with single ejector and flash tank 0.170–0.362 SS with dual ejectors and flash tank 0.234–0.465 [14] Fong et al. 2017 Hong Kong, China Three-story office building 588 ETC 100 SS system LiBr-water 39 Auxiliary heater 0.78 Hybrid system: solar absorption and ground-source radiant cooling [16] Lubis et al. 2016 Indonesia Research center building ETC 181 Single-double effect absorption LiBr-water 239 Auxiliary gas heater 1.4–3.3* COP calculated as cooling capacity divided by natural gas consumption [18] Marc et al. 2012 Reunion Island Four classrooms at university Double-glazed FP 90 SS system LiBr-water 30 Without backup 0.60 [19] Naranjo et al. 2013 Guayaquil, Ecuador 15-story office building 1296 ETC 600 SS system LiBr-water 175 Auxiliary boiler heater 0.60 [20] Xu and Wang 2017 Miami, USA Single zone building CPC 150–275 Variable effect absorption cycle LiBr-water 50 Auxiliary heater 0.88 [21] Agrouaz et al. 2017 Morocco One-story building 200 FP 25 SS absorption system LiBr-water 10 Backup chiller 5 kW 0.19–0.30 This study 2019 Guayaquil, Ec Single zone, office building 450 ETC 24 SS absorption system LiBr-water 15 Without backup 0.60 Manta, Ec 0.53 Puerto Baquerizo, Ec 0.50 Ref. Author Year Location Building type Floor area (m2) Solar collector Collector area (m2) A/C unit Working pair Cooling capacity (kW) Backup system Average thermal COP Comments [13] Abed et al. 2017 Malaysia Single zone building ETC SS with single ejector NH3-H2O 3–5 — 0.127–0.282 SS with single ejector and flash tank 0.170–0.362 SS with dual ejectors and flash tank 0.234–0.465 [14] Fong et al. 2017 Hong Kong, China Three-story office building 588 ETC 100 SS system LiBr-water 39 Auxiliary heater 0.78 Hybrid system: solar absorption and ground-source radiant cooling [16] Lubis et al. 2016 Indonesia Research center building ETC 181 Single-double effect absorption LiBr-water 239 Auxiliary gas heater 1.4–3.3* COP calculated as cooling capacity divided by natural gas consumption [18] Marc et al. 2012 Reunion Island Four classrooms at university Double-glazed FP 90 SS system LiBr-water 30 Without backup 0.60 [19] Naranjo et al. 2013 Guayaquil, Ecuador 15-story office building 1296 ETC 600 SS system LiBr-water 175 Auxiliary boiler heater 0.60 [20] Xu and Wang 2017 Miami, USA Single zone building CPC 150–275 Variable effect absorption cycle LiBr-water 50 Auxiliary heater 0.88 [21] Agrouaz et al. 2017 Morocco One-story building 200 FP 25 SS absorption system LiBr-water 10 Backup chiller 5 kW 0.19–0.30 This study 2019 Guayaquil, Ec Single zone, office building 450 ETC 24 SS absorption system LiBr-water 15 Without backup 0.60 Manta, Ec 0.53 Puerto Baquerizo, Ec 0.50 SS, single -stage absorption cycle; FP, flat plate; ETC, evacuated tube collector; CPC, compound parabolic collector. Open in new tab Table 11 Comparison with literature for absorption cooling systems in tropical climates. Ref. Author Year Location Building type Floor area (m2) Solar collector Collector area (m2) A/C unit Working pair Cooling capacity (kW) Backup system Average thermal COP Comments [13] Abed et al. 2017 Malaysia Single zone building ETC SS with single ejector NH3-H2O 3–5 — 0.127–0.282 SS with single ejector and flash tank 0.170–0.362 SS with dual ejectors and flash tank 0.234–0.465 [14] Fong et al. 2017 Hong Kong, China Three-story office building 588 ETC 100 SS system LiBr-water 39 Auxiliary heater 0.78 Hybrid system: solar absorption and ground-source radiant cooling [16] Lubis et al. 2016 Indonesia Research center building ETC 181 Single-double effect absorption LiBr-water 239 Auxiliary gas heater 1.4–3.3* COP calculated as cooling capacity divided by natural gas consumption [18] Marc et al. 2012 Reunion Island Four classrooms at university Double-glazed FP 90 SS system LiBr-water 30 Without backup 0.60 [19] Naranjo et al. 2013 Guayaquil, Ecuador 15-story office building 1296 ETC 600 SS system LiBr-water 175 Auxiliary boiler heater 0.60 [20] Xu and Wang 2017 Miami, USA Single zone building CPC 150–275 Variable effect absorption cycle LiBr-water 50 Auxiliary heater 0.88 [21] Agrouaz et al. 2017 Morocco One-story building 200 FP 25 SS absorption system LiBr-water 10 Backup chiller 5 kW 0.19–0.30 This study 2019 Guayaquil, Ec Single zone, office building 450 ETC 24 SS absorption system LiBr-water 15 Without backup 0.60 Manta, Ec 0.53 Puerto Baquerizo, Ec 0.50 Ref. Author Year Location Building type Floor area (m2) Solar collector Collector area (m2) A/C unit Working pair Cooling capacity (kW) Backup system Average thermal COP Comments [13] Abed et al. 2017 Malaysia Single zone building ETC SS with single ejector NH3-H2O 3–5 — 0.127–0.282 SS with single ejector and flash tank 0.170–0.362 SS with dual ejectors and flash tank 0.234–0.465 [14] Fong et al. 2017 Hong Kong, China Three-story office building 588 ETC 100 SS system LiBr-water 39 Auxiliary heater 0.78 Hybrid system: solar absorption and ground-source radiant cooling [16] Lubis et al. 2016 Indonesia Research center building ETC 181 Single-double effect absorption LiBr-water 239 Auxiliary gas heater 1.4–3.3* COP calculated as cooling capacity divided by natural gas consumption [18] Marc et al. 2012 Reunion Island Four classrooms at university Double-glazed FP 90 SS system LiBr-water 30 Without backup 0.60 [19] Naranjo et al. 2013 Guayaquil, Ecuador 15-story office building 1296 ETC 600 SS system LiBr-water 175 Auxiliary boiler heater 0.60 [20] Xu and Wang 2017 Miami, USA Single zone building CPC 150–275 Variable effect absorption cycle LiBr-water 50 Auxiliary heater 0.88 [21] Agrouaz et al. 2017 Morocco One-story building 200 FP 25 SS absorption system LiBr-water 10 Backup chiller 5 kW 0.19–0.30 This study 2019 Guayaquil, Ec Single zone, office building 450 ETC 24 SS absorption system LiBr-water 15 Without backup 0.60 Manta, Ec 0.53 Puerto Baquerizo, Ec 0.50 SS, single -stage absorption cycle; FP, flat plate; ETC, evacuated tube collector; CPC, compound parabolic collector. Open in new tab The total annual cooling load is 14 361 kWh, 13 442 kWh and 10 462 kWh in a 450 m3 one-story building located in Guayaquil, Manta and San Cristobal, respectively. In San Cristobal, from June to December the cooling demand decreases significantly. The monthly average collector efficiency ranges from 0.47 to 0.61 in the three cities. Under the conditions of Guayaquil, the collector array has the best performance, in particular between January and April. Similarly, the chiller thermal COP achieves higher values in the warmest months. The monthly averages range from 0.17 to 0.72. In Guayaquil and Manta, the solar COP is in the range of 0.50–0.72. In San Cristobal only on the warmest months, the solar COP reaches a performance in the range of the other two cities; however, from June to December, the solar COP decreases to an average of 0.31. This situation could be attributed to the lower cooling demand presented on these months. Table 11 summarizes some of the previous works on solar absorption cooling applied in tropical or sub-tropical locations. Regarding performance, the system proposed in this study shows good agreement in terms of COP as well as collector area—cooling load ratio; 1.6 m2/kW whereas in the literature range from 0.8 to 4 m2/kW. Although in most cases the solar cooling systems have some type of backup unit, the results of this study as well as the one by Marc et al. [18] show that there is the potential of increased energy savings and greenhouse gas (GHG) emissions mitigation with the application of these types of configurations. Thereby, considering a vapor-compression chiller with the same capacity and a COP of 4, the annual energy savings would range between 2616 and 3590 kWh. Therefore, an annual reduction in GHG emissions about 1120 kg CO2eq is estimated [42]. In economic terms, the proposed system has a lower advantage than the traditional vapor-compression system, due to the subsidies in the electricity tariff (considering a tariff of 0.10 USD/kWh, the annual cost savings would be US$320). 3 CONCLUSIONS The current model has shown that applying a simple strategy control, sound predictions about the thermal performance of a solar absorption cooling system can be obtained. The comparison of the results of the simulations with the measured data shows that the performance of the system is well described by the model developed in TRNSYS. The average error in the operational and performance parameters is 8% during steady-state operation and 21% during transient operation. The larger deviations observed during transient periods (i.e. at start-ups and shutdowns of the system) could be attributed to several factors, including inertial effects that were not considered in some components of the system. The application of a simplified control strategy, although has its limitations in particular when the inlet parameters fall outside the operational range of the system, is an excellent approach in order to size and assess a solar thermal cooling system. The performance of this type of systems depends strongly on several factors that could not be evaluated by applying steady-state methods. The proposed system is capable to provide 100% of the required energy (i.e. no backup system was considered) to air-condition to a 450 m3 one-story building with low internal loads and modeled according to construction standards of the coastal zone of Ecuador. The proposed, which has the following components: 24 m2 ETC field, a 20 kW heat exchanger, a 15 kW single-stage LiBr/H2O absorption chiller, a 35 kW cooling tower and a cold storage tank of 2 m3, could meet up to 90% of the cooling demand of the building considering the climate conditions of Guayaquil, Manta and San Cristobal. The output of the collector and chiller depends on the level of solar radiation, as expected. Months with higher levels of solar radiation are those with higher cooling requirements. Overall, the performance of the proposed system for Guayaquil exceeds the performance of the other two cities. It is important to notice that in tropical locations as the cities evaluated, a collector that can harvest both direct and diffuse radiation is necessary. The operating parameters of the cooling tower significantly influenced the cooling capacity of the system and the outlet temperature of chilled water. Throughout the annual simulation, the temperature of the re-cooling tower was monitored to avoid crystallization problems in the chiller. Although the proposed alternative falls short in economic terms in comparison with the traditional vapor-compression system, due to the subsidies in the national electricity tariff, there is significant potential in terms of climate change mitigation. The latter is an important factor to take into consideration, especially if there is a serious commitment to move toward more sustainable urbanization, and in the case of Galapagos Islands, this matter becomes even more relevant. REFERENCES 1. WMO . 2018 . WMO Statement on the status of the global climate in 2017 [Internet]. World Meteorological Organization . Geneva : World Meteorological Organization . WMO-No. 1108. http://www.wmo.int/pages/mediacentre/press_releases/documents/WMO_1108_EN_web_000.pdf . Google Preview WorldCat COPAC 2. Davis LW , Gertler PJ . Contribution of air conditioning adoption to future energy use under global warming . Proc Natl Acad Sci USA 2015 ; 112 : 5962 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 3. Santamouris M . Cooling the buildings – past, present and future . Energy Build 2016 ; 128 : 617 – 38 . Google Scholar Crossref Search ADS WorldCat 4. Eicker U , Pietruschka D , Schmitt A , Haag M . Comparison of photovoltaic and solar thermal cooling systems for office buildings in different climates . Sol Energy 2015 ; 118 : 243 – 55 . Google Scholar Crossref Search ADS WorldCat 5. Fan Y , Luo L , Souyri B . Review of solar sorption refrigeration technologies: Development and applications . Renew Sustain Energy Rev 2007 ; 11 : 1758 – 75 . Google Scholar Crossref Search ADS WorldCat 6. Eicker U , Pietruschka D , Haag M , Schmitt A . Energy and economic performance of solar cooling systems world wide . Energy Procedia 2014 ; 57 : 2581 – 9 . Google Scholar Crossref Search ADS WorldCat 7. Henning H-M . Solar assisted air conditioning of buildings – an overview . Appl Therm Eng 2007 ; 27 : 1734 – 49 . Google Scholar Crossref Search ADS WorldCat 8. Hassan HZ , Mohamad AA . A review on solar-powered closed physisorption cooling systems . Renew Sustain Energy Rev 2012 ; 16 : 2516 – 38 . Google Scholar Crossref Search ADS WorldCat 9. Abed AM , Alghoul MA , Sopian K et al. Enhancement aspects of single stage absorption cooling cycle: a detailed review . Renew Sustain Energy Rev 2017 ; 77 : 1010 – 45 . Google Scholar Crossref Search ADS WorldCat 10. Baldwin C , Cruickshank CA . A review of solar cooling technologies for residential applications in Canada . Energy Procedia 2012 ; 30 : 495 – 504 . Google Scholar Crossref Search ADS WorldCat 11. Bataineh K , Taamneh Y . Review and recent improvements of solar sorption cooling systems . Energy Build 2016 ; 128 : 22 – 37 . Google Scholar Crossref Search ADS WorldCat 12. Pietruschka D . 2010 . Model Based Control Optimisation of Renewable Energy Based HVAC Systems . De Montfort University Leicester . Google Preview WorldCat COPAC 13. Abed AM , Sopian K , Alghoul MA et al. Experimental evaluation of single stage ejector-absorption cooling cycle under different design configurations . Sol Energy 2017 ; 155 : 130 – 41 . Google Scholar Crossref Search ADS WorldCat 14. Fong KF , Lee CK , Zhao TF . Effective design and operation strategy of renewable cooling and heating system for building application in hot-humid climate . Sol Energy 2017 ; 143 : 1 – 9 . Google Scholar Crossref Search ADS WorldCat 15. Jaruwongwittaya T , Chen G . A review: Renewable energy with absorption chillers in Thailand . Renew Sustain Energy Rev 2010 ; 14 : 1437 – 44 . Google Scholar Crossref Search ADS WorldCat 16. Lubis A , Jeong J , Saito K et al. Solar-assisted single-double-effect absorption chiller for use in Asian tropical climates . Renew Energy 2016 ; 99 : 825 – 35 . Google Scholar Crossref Search ADS WorldCat 17. Lubis A , Jeong J , Giannetti N et al. Operation performance enhancement of single-double-effect absorption chiller . Appl Energy 2018 ; 219 : 299 – 311 . Google Scholar Crossref Search ADS WorldCat 18. Marc O , Anies G , Lucas F , Castaing-Lasvignottes J . Assessing performance and controlling operating conditions of a solar driven absorption chiller using simplified numerical models . Sol Energy 2012 ; 86 : 2231 – 9 . Google Scholar Crossref Search ADS WorldCat 19. Naranjo-Mendoza C , Rousse DR , Quesada G . Modeling of a solar absoprtion cooling system for Guayaquil, Ecuador . Int Conf Renew Energy Res Appl 2013 ; 2013 : 853 – 6 . WorldCat 20. Xu ZY , Wang RZ . Simulation of solar cooling system based on variable effect LiBr-water absorption chiller . Renew Energy 2017 ; 113 : 907 – 14 . Google Scholar Crossref Search ADS WorldCat 21. Agrouaz Y , Bouhal T , Allouhi A et al. Energy and parametric analysis of solar absorption cooling systems in various Moroccan climates . Case Stud Therm Eng 2017 Mar ; 9 : 28 – 39 . Google Scholar Crossref Search ADS WorldCat 22. Zhai XQ , Qu M , Li Y , Wang RZ . A review for research and new design options of solar absorption cooling systems . Renew Sustain Energy Rev 2011 ; 15 : 4416 – 23 . Google Scholar Crossref Search ADS WorldCat 23. Aliane A , Abboudi S , Seladji C , Guendouz B . An illustrated review on solar absorption cooling experimental studies . Renew Sustain Energy Rev 2016 ; 65 : 443 – 58 . Google Scholar Crossref Search ADS WorldCat 24. Liu L , Li Z , Jing Y , Lv S . Energetic, economic and environmental study of cooling capacity for absorption subsystem in solar absorption-subcooled compression hybrid cooling system based on data of entire working period . Energy Convers Manag 2018 ; 167 : 165 – 75 . Google Scholar Crossref Search ADS WorldCat 25. El Telegrafo . 2019 . Turismo a las Islas Galápagos Superó los 275.000 Visitantes en 2018 . https://www.eltelegrafo.com.ec/noticias/ecuador/1/turismo-galapagos-2018 (31 Jan 2019, date last accessed) . 26. ARCONEL . 2018 . Estadísticas del Sector Eléctrico: Producción . http://www.regulacionelectrica.gob.ec/estadistica-del-sector-electrico/produccion/ (25 May 2018, date last accessed) . 27. ARCONEL . 2018 . Estadísticas del Sector Eléctrico: Demanda . http://www.regulacionelectrica.gob.ec/estadistica-del-sector-electrico/demanda-mensual/ (25 May 2018, date last accessed) . 28. IEconomics . 2018 . Ecuador Gross Average Nominal Monthly Wages . https://tradingeconomics.com/ecuador/wages (25 May 2018, date last accessed) . 29. CIE . 2008 . Atlas Solar del Ecuador . Quito . www.energia.org.ec. (5 May 2012, date last accessed) . Google Preview WorldCat COPAC 30. Castro M . 2011 . Hacia Una Matriz Energética Diversificada en Ecuador . Quito . www.ceda.org.ec. (17 July 2018, date last accessed) . Google Preview WorldCat COPAC 31. Agyenim F , Knight I , Rhodes M . Design and experimental testing of the performance of an outdoor LiBr/H2O solar thermal absorption cooling system with a cold store . Sol Energy 2010 ; 84 : 735 – 44 . Google Scholar Crossref Search ADS WorldCat 32. SOLAIR . 2009 . Training course on solar cooling. Chapter C: predesign - system sizing . Intelligent Energy - Europe (IEE) . 27 . Google Preview WorldCat COPAC 33. Römmling U , Albers J . 2002 . TRNSYS Type 107 Part Load Simulation of Single Staged Absorption Chillers in Quasi Steady States . Berlin : Technische Universität Berlin . (21 May 2018, date last accessed) Google Preview WorldCat COPAC 34. SEL . 2006 . Trnsys 16 Manual , Vol. 1 . Madison . http://sel.me.wisc.edu/trnsys/ (1 June 2012, date last accessed) Google Preview WorldCat COPAC 35. Kingspan Renewables Limited . THERMOMAX Technical Design Guide . Bangor . www.kingspansolar.com (15 June 2012, date last accessed) . Google Preview WorldCat COPAC 36. Ziegler F , Hellman H , Schweigler C . 1999 . An approximative method for modelling the operating characteristics of advanced absorption chillers . In 20th International Congress of Refrigeration . Sydney . International Institute of Refrigeration Google Preview WorldCat COPAC 37. Labus J . 2011 . Modelling of Small Capacity Absorption Chillers Driven by Solar Thermal Energy or Waste Heat . Universitat Rovira I Virgili . Google Preview WorldCat COPAC 38. Albers J . 2011 . Type 177: Absorption Chiller - Characteristic Equation Model . Technische Universitat Berlin 39. Braun JE . 1988 . Methodologies for the Design and Control of Central Cooling Plants . University of Wisconsin . Google Preview WorldCat COPAC 40. Meteotest . 2012 . Meteonorm Handbook Part II: Theory . www.meteonorm.com. (8 June 2012, date last accessed) . 41. Jähnig D , Klein F . 2010 . Package Solution Description , Vol. 17 . Gleisforf . Google Preview WorldCat COPAC 42. Ramírez A , Rivela B , Melendres AM , Espinoza S . 2015 . Life cycle assessment of Ecuadorian electricity . Lima, Perú . Google Preview WorldCat COPAC © The Author(s) 2020. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com
TI - Modeling and simulation of a small-scale solar-powered absorption cooling system in three cities with a tropical climate
JF - International Journal of Low-Carbon Technologies
DO - 10.1093/ijlct/ctz040
DA - 2020-02-20
UR - https://www.deepdyve.com/lp/oxford-university-press/modeling-and-simulation-of-a-small-scale-solar-powered-absorption-rbDcSGLEQq
SP - 1
VL - 15
IS - 1
DP - DeepDyve
ER -