K. T. Witte1*, J. Albers2, M. Krause3, M. Safarik4, F. Besana1, W. Sparber1

1 European Academy Bolzano, Institute of Renewable Energy, Viale Druso 1, 39100 Bolzano, Italia
2 Technische Universitat Berlin, Institut fur Energietechnik KT-2, Marchstrabe 18, 10587 Berlin, Germany
3 Fraunhofer Institute for Building Physics, Gottschalkstr. 28a, 34127 Kassel, Germany
4 Institute of Air Conditioning and Refrigeration (ILK), Bertolt-Brecht-Allee 20, 01309 Dresden, Germany

* Corresponding Author, kwitte@eurac. edu.

Abstract

In order to propagate the installation of solar cooling systems dimensioning and system design have to be optimised. Although the transient simulation software TRNSYS is well established for this purpose, considerable deviations to measured values are possible, when components are not correctly used or designed for the simulation. For this work two different models (so called Types) are investigated in parallel to simulate the part load behaviour of the absorption chiller EAW Wegracal SE 15. The paper reports on the possi­bilities and application limits for both TRNSYS Types. Their differences in accuracy regarding laboratory test data as well as real time data of a solar cooling system are dis­cussed. The basic methodology as well as the applied modifications to adapt the TRNSYS — Types to the EAW chiller are described. Finally the necessity to improve both models for non-nominal flow rates is shown.

Keywords: Solar cooling, EAW absorption chiller, Simulation, Characteristic Equation

1. Introduction

Solar Combi-Plus systems use solar energy to provide heat for domestic hot water and heating in winter as well as for domestic hot water and cold production in summer. Thus high solar fractions are possible over the year. Moreover such systems contribute to decrease the peak power demand during summer, which is partially caused by the growing number of small electrically driven air­conditioning appliances in the sunny and hot southern countries of Europe [1]. For the optimum design of solar cooling systems computer simulations are essential. To validate these simulations real time data of Solar Combi-Plus systems are required. Therefore, the European Academy (EURAC) in Bolzano is monitoring one of them at the premises of Ebner Solartechnik, using the absorption chiller EAW Wegracal SE 15 [2]. For the simulation of this plant and its utilisation the simulation software TRNSYS [3] is used. In a first step the available component of the ‘standard package’ of TRNSYS Version 16.1 was chosen to simulate the absorption chiller (Type 107). Since the implemented control strategy of this Type as well as the performance data differ from real operation of the EAW Wegracal SE 15 chiller, both the source code and the performance data of Type 107 have been modified. In a second step, another model based on the method of characteristic equations [4] (Type 177) was adapted to the EAW chiller and the simulation results have been compared to the same measurements.

2. Simulation models

A solar air-conditioning pilot plant has been built and is currently running in PROMES laboratory (Perpignan France). This pilot, of a daily cooling capacity of 20 kWh, is based on a solid/gas thermochemical sorption process that is powered at 60-70°C by 20m2 of flat plate solar collectors. The thermochemical sorption process rests on the coupling of a reversible chemical reaction between a reactive solid (BaCl2) and a liquid/gas phase change of a refrigerant (NH3). Its functioning mode is intrinsically discontinuous and cyclic and is thus well adapted to the storage/transformation of solar energy. Indeed, the process comprises two main reactive phases that are carried out under different thermodynamic constraints: a diurnal regeneration phase followed by the useful production of cold during the night phase. During the first phase, the heat supplied to the reactor in the range of 50 to 70°C by the flat plate solar collectors enable the decomposition reaction of the reactive salt which releases the reactive gas that condenses in a condenser and accumulated at ambient temperature. When there is a surplus of solar energy, this one can be stored at 80°C by melting a phase change material (wax) for either enabling a faster heating of the reactor during the morning, or for domestic hot water production The cold production during the night consists in cooling the reactor which reabsorbs chemically the reactive gas that evaporates in the evaporator and thus producing the cold. The produced cold at -5°C to 0°C is stored in another PCM that solidifies below 4°C, in order to be delivered according to the cooling demand all along the day. An analysis of experimental results leads to an averaged efficiency of 50% for the solar collectors and a process COP ranging from 30 to 40%, leading thus to a daily cooling productivity for the thermochemical process of about 0.8 to 1.25 kWh of cold at 4°C per m2 of solar collector.

3. Results

3.1 Expected results

The investment cost of solar cooling systems is much higher than those of conventional air conditioning equipments. For instance, the cost of the sorption installations vary widely depending on the facilities, from 4000 €/kW up to 10 000€/kW, while the price for conventional compression air conditioning is around 300 €/kW. On the other hand, even if the technologies involved are common they are still poorly understood when set up in a building. Before considering a wide dissemination of these processes, it is necessary to fill a number of gaps concerning the estimation of energy, environmental and financial efficiency, the optimization of the thermodynamic processes and the lack of decision-making tools for pre studies and studies. To allow wider dissemination of solar cooling, the issues of research programs for the years to come are therefore

to reduce installation costs and to ensure optimum performance. The first expected result for ORASOL is to define what is the best cooling process according the building type and the climate. This answer will be brought by the analysis carried out by the pool 1. Then it is important to provide accurate but also validated simulation software for each solar cooling technology. These tools aim to support all research works on solar cooling and must also be used as a basis for more sophisticated ones as, for instance, sizing or optimization tools.

On this day (07/07/2008) no cooling energy was needed in the rooms because the ambient air temperature was approximately 20 °C. Hence, the ice-storage could be charged. At the beginning of the day the storage was partially charged with approximately 15 kWh of cooling capacity. At the end of the day the storage was completely iced and a cooling capacity of 35 kWh had been available. This cooling energy had been used in the morning hours of the next days. Figure 6 shows the power supplied to the generator and the evaporator, and the solar radiation on this day.

The solar radiation on this day was very changeable. This is the reason why the operation of the chiller did not start before 11:30. It is remarkable that in spite of the inconstant solar radiation the power of the evaporator is very constant. The average COP on this day was 0.50. Figure 7 shows the operating temperatures of the chiller.

5. Conclusions

The developed absorption chiller is able to cool the rooms where the cooling system is installed, and has still some buffer capacities for higher cooling loads. In the mornings (until 9:45) and in the evenings (after 16:30) cooling performance is lacking and has to be covered with the ice storage. The ice storage could be charged on days when no cooling was required. Even on days with changeable solar radiation the charging of the ice-storage is possible. Despite the promising results so far, additional work is required. For example, the automatic control of the system must be upgraded and more measurement data have to be acquired to optimize the complete system.

References

[1] Zetzsche, Koller, Brendel, Muller-Steinhagen: Solar cooling with an ammonia/water absorption chiller, 2nd International Conference Solar Air-Conditioning, Oct. 18th/19 th 2007, Ostbayerisches Technologie-Transfer-Institut e. V., Regensburg, Germany, p. 536-541 [14]

W. Mittelbach, T. Btittner and R. Herrmann

SorTech AG, Weinbergweg 23, 06120 Halle, Germany, walter. mittelbach@sortech. de

ABSTRACT

Beside the well-known liquid absorption chillers with LiBr-water as sorption pair also solid adsorption chillers may be used for generating chilled water with low-temperature heat. In the range of cooling capacities of 70 kW and higher Japanese products are on the market for many years using silica gel as adsorbent and water as refrigerant. But due to the much larger volume and weight of the machines compared to LiBr absorption chillers adsorption chillers only found niche markets up to now. The main reason for the large volume of the adsorption chillers sold up to now is the low heat transfer rate between silica gel and the heat exchanger surfaces. SorTech AG developed a coating process, which enables to apply heat exchanger surfaces directly with the adsorbent. This enabled the development of more compact and lightweight adsorption chillers. Based on this technology SorTech AG developed a small adsorption chiller with a nominal cooling capacity of 7.5 kW. The objective is to provide a compact machine, which may be used for solar air-conditioning in private homes and small offices.

Keywords: adsorption chiller, solar cooling, silica gel

Victor Afyan1 and Arsen Karapetyan2

1 SolarEn, LLC, 2/2 Shrjanayin St., Yerevan 0068, Armenia.

2 SolarEn, LLC, 2/2 Shrjanayin St., Yerevan 0068, Armenia.

* Corresponding Author, victor afyan@solaren. com

Abstract

10 kW PV power station is integrated into the roof of the five-store medical center building and connected to the 3-phase grid on the net metering basis. For cooling needs of the building there is a hot water driven lithium bromide absorption chiller with 420kW cooling capacity. The chiller is powered by a gas boiler and supplies chilled water to the fan-coils system of the building. PV station is installed and tested. Absorption chiller is installed and commissioned. Technical data for PV and cooling systems are presented.

Keywords: BIPV, net metering, cooling, absorption chiller.

1. Introduction

Building integrated photovoltaic (BIPV) and grid-tied solar power stations are well known and implemented in developed countries due to the favorable feed-in tariff and rebates. Although PV technology has a long history in CIS countries, particularly for autonomous power supply applications, its BIPV implementation is hampered by absence of grid connection legal mechanisms and financial stimuli.

PV power plant as well as other renewable energy (RE) power plants connection and parallel operation with the grid has been ensured and regulated by the net metering mechanism adopted in Armenia in 2005. According to the regulation all RE power plants and even combined heat and power (CHP) units with capacity up to 150kW can benefit parallel operation with the grid based on the net metering.

The PV power station and tri-generation energy efficient system has been designed for the renovated building of the Armenian-American Wellness Center in Yerevan. The design is based on 10kW roof-integrated grid-connected PV power station, hot water driven lithium bromide absorption chiller with 420kW cooling capacity and cooling tower, gas genset with 110kWe and 150 kWth capacity and a back-up gas boiler. The CHP issue is still pending but other components of the system are already installed and described below

P. Kohlenbach*, D. Rossington and A. Weigand

1 CSIRO Energy Technology, PO Box 330, Newcastle, NSW, 2300, Australia
Corresponding Author, paul. kohlenbach@csiro. au

Abstract

CSIRO Energy Technology is developing a small-scale desiccant-based air-conditioning system for residential applications. In this context, a desiccant wheel made of a novel material has been experimentally tested for its dehumidification performance. The material is an iron-alumino-phosphate zeolite with an AFI structure and traded under the name of FAM Z-01. A 300mm diameter desiccant wheel was tested under varying inlet conditions of temperature and humidity with regard to its dehumidification performance. It was found that for constant regeneration humidity the maximum moisture removal capacity of the material is 17 grams of water per kg dry air at 50°C regeneration temperature and 24 grams of water per kg dry air at 80°C regeneration temperature from an inlet air stream of 40 °C and 95% relative humidity. At supply inlet temperature between 10 and 30°C and supply inlet relative humidity between 20 and 50% it was found that the difference in moisture removal at a regeneration temperature of 50 °C and at 80 °C is around 1 g/kg d. a.. At varying regeneration humidity (matching ambient conditions) it was found that the moisture removal is considerably lower, even though the regeneration air is supplied at the same temperature. Maximum moisture removal was 5.1 g/kg d. a. and 14.5 g/kg d. a. for supply inlet conditions of 40°C/95% RH at 50 degC and 80 degC regeneration temperature, respectively.

Keywords: Dehumidification, desiccant wheel, zeolite, FAM Z-01

1. Introduction

CSIRO is currently developing a solar-powered air-conditioning unit for residential houses, using a desiccant-evaporative process to provide cool and dehumidified air. This process is very well suited for the recovery of low-grade solar energy or waste heat. Thermal energy and water are used to provide air-conditioning, hence consuming only a very small amount of electrical power. As part of the development CSIRO is testing novel desiccant wheel materials for dehumidification purposes. The two most common materials for desiccant wheels are silica gel and LiCl due to their low cost and ease of handling. They are however limited in their moisture removal capacity, especially at regeneration temperatures below 80 degC. Recently researchers and manufacturers have been developing advanced materials to increase the moisture removal capacity and hence to allow for smaller unit size.

Jia et. al. [1, 2] describe a comparison between a novel composite desiccant wheel made of silica gel and lithium chloride and a conventional wheel made of silica gel only. They found that the composite wheel has a greater moisture removal capacity compared to the silica gel wheel, especially at lower inlet humidity. The regeneration temperature of the composite wheel was found to be lower than that of the pure silica gel wheel. Tokarev et. al. [3] have analysed a composite sorbent based on CaCl2 as an
impregnated salt and MCM-41 as a host matrix. At regeneration temperatures between 70 and 120 degC the moisture removal capacity of the composite was greater than of silica gel. Cui et. al. [4] investigated the properties of DH5, DH7 and 13x adsorbents with regard to their use in desiccant cooling systems. Their results show that DH5 and DH7 adsorbents have greater moisture removal capacity than silica gel when tested at a regeneration temperature of 100 degC. Restuccia et. al [5] also investigated a composite sorbent SWS-1L, a mesoporous silica gel KSK impregnated with CaCl2. It showed a promising moisture removal capacity of up to 0.7 g of water per 1 g of dry sorbent at regeneration temperatures of 90-100°C. Kakiuchi [6] presented the FAM-Z01 material used in this work as an application for adsorption heat pumps and proposed the application for desiccant wheels. This application was further investigated by Oshima et. al [7] who evaluated the performance of a desiccant rotor containing FAM-Z02 zeolite material. Various regeneration temperatures and air inlet conditions have been investigated in a parameter study. The moisture removal of the FAM-Z02 rotor was found to be 11-22% higher than that of a silica gel rotor at regeneration temperatures of 50-70°C.

One important aspect of using adsorbents in a solar cooling system is the long-term stability of the desiccant. Earlier investigations by Belding et. al [8] have shown that silica gel and 13x adsorbents can lose up to 63% and 13%, respectively, of their original moisture removal capacity after 50,000 cycles. The FAM Z-01 material used in this work has been tested by the manufacturer and has shown a 5% capacity loss after 50,000 cycles [6].

2. System and methodology

The experimental testing has been undertaken at CSIRO’s Energy Technology site in Newcastle, Australia. Figures 1 and 2 show the test rig used for experimental purposes.

As shown in Figure 1 and 2, the test rig consists of two separate conditioned air streams, one for desiccant wheel supply, and the other for desiccant wheel regeneration. Each of these two air streams enter via an intake filter (item 1 of Figure 1), and is then pressurised by a centrifugal fan (item 2) which is controlled by a variable speed drive to enable air volume control. Peak flow of 1000m3/hr is achievable with a 300Pa pressure drop across the desiccant wheel. The air stream then passes through two cooling coils. The first coil (item 3) is cooled with a 2°C chilled glycol solution capable of dehumidification of the leaving air stream to a moisture ratio of approximately 7 g/kg dry air. The second cooling coil (item 4) is cooled with a -5°C chilled glycol solution capable of further dehumidification of the leaving air stream to a moisture ratio of approximately 4 g/kg dry air. The dehumidified air stream then passes through a primary electric heater bank (item 5). This heater bank is capable of heating the air stream to 90°C in the case of the regeneration air stream, and 40°C in the case of the supply air stream. The air stream is then humidified as required using a steam injection lance (item 6). Low pressure steam at 1.5 bar is injected in the air stream via nozzles at a rate of up to 45 g/kg dry air. Finally the air stream passes though a secondary electric heater bank (item 7). The secondary heater bank is of similar capacity as the primary heater bank allowing for load sharing and fine temperature control. The supply and regeneration air streams are then ducted to the test desiccant wheel. This can be done in counter flow and parallel flow arrangement. The supply and regeneration air streams leaving the desiccant wheel are ducted from the wheel and exhausted outside. Temperature and Humidity are measured and recorded after each of the control elements described above. The temperature and humidity entering and exiting the desiccant wheel is calculated by averaging a number of sensors distributed over the cross section of the ductwork entering and leaving the desiccant wheel. Volume flow rate of the supply and regeneration air streams is calculated from velocity measured at the entering side of the desiccant wheel.

The basic control schemes have been applied on the following three basic configurations for the installation:

2.2. Direct use configuration

The basic scheme used to execute the simulations in this case is the one on the following figure.

Fig.2. Configuration for direct use

This corresponds to the habitual configuration for installations with absorption chiller. In this case to start the chiller working they must be fulfilled two requirements: to have solar energy enough stored on the tank and to have cold demand.

The main results within the first level are the

… primary energy ratio of the installed solar assisted heating and cooling ((Equation 1)

. primary energy ratio of an assumed reference system working under the conditions as the installed SHC system ((Equation 2).

Herewith the annual savings of primary energy through the utilisation of the SHC system can be expressed.

Further through the calculated cost per installed cooling capacity ((Equation 4), it will be possible in the following years to draw a learning curve for SCH systems.

3.1. Expected results — 2nd level

The main result within the second level is the calculation of the solar thermal energy which could not be exploited by the SHC system — this shows the “efficiency” of the SHC system ((Equation 12 and (Equation 13, see Fig. 5)

Pedro Adao1*, Manuel Collares Pereira1, Andrea Costa2

1 Ao Sol SA (SunCool SA), Lugar da Sesmaria Limpa, 2135-402 Samora Correia, Portugal
2 ACE, Pfalzerstr. 75, 83109 Grosskarolinenfeld, Germany
* Corresponding Author, pedro. adao@aosol. pt

Abstract

This work presents the development results of a novel absorption chiller for solar cooling applications built at Ao Sol (SunCool laboratories) in Portugal. The chiller is at pre­commercial stage at the time of writing. Focus of the ongoing development work is the optimisation of technical and economic parameters such as overall dimensions. A thermo­economic analysis will give insight on the commercial potential and challenges.

Under present conditions, dictated by a new European legislation and current fuel price, the use of small absorption chillers coupled to solar thermal technology for residential applications turns out to be potentially both environmentally sound and economically viable. Necessary requirements are cost-efficient solar collectors and chillers, and an economically optimized combined system.

Relying on previous work performed at the University of Lisbon, a compact absorption chiller working with a solution of ammonia and water has been built and tested. The device is mainly made of plate heat exchangers and it is directly cooled by air, thus avoiding the need of a wet cooling tower. The control strategy is aimed to an overall concept that includes collector field, storage, gas back-up, chiller and building.

The thermodynamic design is presented along with experimental results of the thermal performance — in terms of chilled water production and coefficient of performance. The economic impact of the developed chiller for a common residential application in Portugal is compared with a reference system and discussed.

Keywords: solar cooling, air-cooled chiller, ammonia-water absorption, net present value

1. Introduction

The present European Directive on Energy establishes ambitious goals for the year 2020: 20% contribution of renewable energies to the final energy demand and 20% reduction of the final energy consumption through energy efficiency measures [1]. This creates a very strong drive for the solar penetration in the Heating and Cooling of Buildings, responsible for about 40% of the total final energy consumption in the EU [1].

Combining this directive with the ESTIF solar thermal installed capacity goals for 2020 [2 ESTIF Newsletter July 2007] an upper limit of about 4 million new solar driven machines can be anticipated for the residential and services market.

If the more ambitious ESTIF expectations are taken into consideration this figure will be multiplied by a factor of 5 to 10 to 2030.

The present situation in Europe shows a huge potential for renewable energies. In the building sector, with its highest emissions share, the potential for energy savings is over proportional high. Building cooling is rising to become a crucial issue in the next years. Solar energy is indisputably assessed as a major contributor to these savings and it is being fully backed by policy makers through long-term directives at EU-level.

Trends in Southern Europe: Southern European countries are in a particularly dire situation with a growth of energy consumption in the buildings sector precisely because of the widespread use of electrically-driven air-conditioning units; a situation which will even worsen with the warming trends associated with global warming. In the Portuguese residential and services sector the average increase rate of electricity consumption over the last three years is more than 4 times the average rate of the European Union [3]. This is mainly due to the rise in air-conditioning demand traditionally achieved with highly inefficient small-size chillers (window units). This generates a tendency, which is against the goal of 20% energy use reduction and requires even stronger action than in Northern European countries.

In the past, INETI and some of the people involved in the present development participated in a European funded project [4] to analyze the technical and economic potential of solar absorption cooling in the Mediterranean region. Essentially the conclusion was that — through savings induced by solar energy — an all-year system providing heating, cooling and domestic hot water would compare favourably with a conventional system with a boiler for heating and an electric chiller for cooling with a payback time of less than 10 years. The conclusions and ideas developed in that study were truly influential on the decision of AO SOL to take up the present R&D effort.

The exergy-topological method also called exergy graph method is a new approach in exergy studies based on the arrangement or mapping of elements (links and nodes) of a network. In this method, the components: turbines, pumps, heat exchangers, etc linked by the pipes, valves and other devices are the major elements. Different steps involved in the application of the topological methodology are summarized below:

■ Step 1: drawing of the flow sheet of the system,

■ Step 2: establishment of the exergy flow graph,

■ Step 3: determination of the matrix of incidence,

■ Step 4: determination of the flow parameters,

■ Step 5: determination of the thermodynamic characteristics and

■ Step 6: analysis of the results.

The useful parameters for the analysis of the system are as follows:

V = E°ut / E” (Degree of thermodynamic perfection of i th-element) (1)

l = Ei” — E°ut (Exergy loss of i th-element) (2)

rfx = Ei /Ea (Exergy efficiency of i th-element) (3)

в = Ef / El (Coefficient of influence of i th-element) (4)

Vj, = ElU / El” (Degree of thermodynamic perfection of the system) (5)

n

= ^ili (Total exergy loss of the system) (6)

i=1

Пі = Ej / Ej (Overall exergy efficiency of the system) (7)

Where, E“ and E°ut are the sum of exergy flows at the inlet and outlet of the element i; Eut and Ef are the sum of useful and available exergy of i th-element and the subscript E refers to the system. The exergy rate of a j-flow is given by

Ej = mej (8)

ej is the specific exergy flow and m represents the mass flow rate. Specific exergy can be calculated by the following equation for any flow [8]:

ej = (hj — h0) -T0(Sj — s0) + ef (9)

Where hj and sj are specific enthalpy and specific entropy at the point under consideration. T0, h0 and s0 are the temperature, specific enthalpy and entropy at the restricted dead conditions. ef is the chemical exergy.

The specific exergy transfers by heat and by work are respectively given by Eqs. 10 and 11.

ej = q} (1 — TJ Tj) (10)

ej = Wj (11)

Where qj is the specific amount of heat, Tj is the average thermodynamic temperature of the

working body at which heat is added or removed, w3 is the specific work.

The methodology of calculation of useful and available exergy rates is described in Ref.[8]. A description and exemplification of this approach can be found in Mago et al. [4] and Nikulshin et al. [5-7].