Chapter 2

Literatures Review on Integrating Absorption Cooling Systems to Thermal Electrical Power Generation Plants
2-1 Overview:
The purpose of this chapter is to review the published researches and studies done on absorption cooling systems, especially, those using Li Br H2O pair as a working fluid. Its a trial to explore the different aspects of Integrating Absorption Cooling Systems to Thermal Electrical Power Generation Plants and its essential and potential role in improving the overall efficiency, economics, and

environmental impacts of existing and future conventional electricity generation plants.

Many researches have been conducted and are being conducted to enhance the performance of absorption cooling cycle, some are interested in overcoming the well known limitations by using new additives. Others investigate the optimum performance conditions like gas consumption rate, solution circulation rate, the flow rate effect on the distribution quality and falling film heat transfer coefficien t, and the geometrical details through modeling and simulation programs for the cycle different components. In addition, studies for absorption cycle energy, mass, and salt balances, and an overall heat transfer coefficient have been performed. Other investigators are studying the economic, environmental and thermal impacts of the integration of absorption cycles with a power generation system. The following survey will focus on three main subjects; Performance enhancement of Gas Turbines by Inlet Air Cooling (GTIAC) using absorption cooling systems, recent improvements in absorption cooling cycle focusing on economic and environmental analysis of electricity generation projects, and finally, the modeling and simulation of absorption cooling cycles utilizing (H2O/ Li Br) as a working pair fluid .

2-2

Performance enhancement of Gas Turbines by Inlet Air Cooling (GTIAC) using absorption cooling systems

Absorption cooling is a mature technology with the first machine developed in 1859 by Ferdinand Carre. For the closed cycle process, a binary working fluid that consists of the refrigerant and an absorbent is necessary. Carre used the working fluid ammonia/water (NH3 / H2O). Today the working pair lithium bromide as absorbent and water as the refrigerant is most commonly used for building climatization (H2O/ Li Br). In contrast to the ammonia/water system with its pressure levels above ambient pressure, the water/lithium bromide absorption cooling machine(ACM) works under vacuum because of the low vapour pressure of the refrigerant water. In 1945, the company Carrier Corp, US, developed and introduced the first large commercial single-effect ACM using water/lithium bromide with a cooling power of 523 kW. In 1964, the company Kawasaki Heavy Industry Co, Japan produced the first double-effect (DE) water/lithium bromide ACM (Hartmann, 1992). Dickson (1975) indicated that the inlet air cooling system must be designed to avoid icing at the compressor inlet or anywhere in the air intake structure. Ice fragments sucked into the compressor can cause serious structural damage. Icing is a potential problem anytime the ambient air temperature drops near the freezing mark. Compressor bleed air can be used to internally heat compressor inlet surfaces or directly injected into the inlet air stream.

Wagar (1980) has routed back the exhaust from the turbine through the inlet air structure within closed heat exchanger surfaces, and electric heat tracing has been used as well. The potential icing problem is exacerbated for inlet air cooling systems because warm ambient air will almost always be saturated after passing through the inlet air cooling coils. When the air is drawn into the mouth of the compressor, its velocity increases and its temperature drops further as air enthalpy is transformed into kinetic energy in an adiabatic process. Condensate icing can occur if the temperature drops below freezing. A temperature drop of about 10 F

is common. Therefore, the design inlet air temperature should be at least 42 F to avoid potential icing problems.

One of the most important applications of Absorption Refrigeration system is the. This new power enhancing technique was investigated and modeled by

Kohlenberger (1995). The study based on Westinghouse 501 D5 open cycle combustion turbine of 118.5 MW nominal ISO base rating and 10023 btu/kWh and a refrigerated GTIAC system with the inlet air cooled to 4.4 oC . Combining the absorption system to the power cycle has led to increase the net power output by 24% , gross heat rate is reduced by 6%, and exhaust gas temperature is reduced 5% .The study concludes to the fact that, GTIAC system provides very attractive economic paybacks in addition to the significant energy improvements of gas turbine power systems.

Meloche et al (1996) assumes a heat supply recovery from the heat recovery boiler at (100 oC) and the heat return back to the boiler at (75 oC). The gas turbine has an electrical efficiency of 35%, heat efficiency of 53%, and 12% losses. Meloche examined running the gas turbine at an input of 100 units of fuel and determines the maximum cooling using a compression chiller and an absorption unit. The calculation is based on maximum chilled water production from all the available energy, which is also a reason for the electricity to chillers to be large for the absorption chiller. The absorption chillers are being supplemented with electrical chillers from the excess electrical output. The paper concludes to the result that there is no difference in the economics for using either heat driven or compression chillers for a new CHP plant.

Schweigler et al (1996) describe a single-effect, half-effect combination unit [SE/HE]. A single-effect machine can operate at (80 oC), but in district heating networks the summer temperature for hot water is in the range from (70-80 oC). Thus, a combination SE/HE will allow use in district heating networks. Two advantages of the SE/HE chiller are a reduction in heat exchanger area and it allows for a temperature glide in the generator of (30 oC). The SE/HE chiller operates with

a COP around 0.55 to 0.6 with driving heat supply / return temperature of (90/60
o

C). The operation of the SE/HE chiller is driven by the available heat source

temperature and flow rate. If the heat source temperature is above (80 oC) then the SE/HE acts as a SE unit with a COP of 0.7. Any temperatures below this the SE/HE acts as a combination SE/HE or pure HE, with the lower bound on the COP being 0.35 . The main conclusion from the pilot plant test are; operating costs are almost the same as those of the theoretical performance predictions. In part load operation, COP is increased while temp. glide in the hot water remained constant. An increase in cooling capacity was achieved per unit of hot water mass flow compared with a standard single effect unit.

Ma et al (1996) examined a 100 ton (350 kW) half-effect absorption chiller driven by low temperature waste heat from a combined heat and power (CHP) plant in Beijing, China. The half-effect chiller is able to operate at a hot water temperature of (86oC) to produce (9 oC) chilled water using (32 oC) cooling water at a COP of 0.40 . A single-effect chiller can not produce the specified chilled water with the given heat source temperature so a half-effect device must be used.

The integration of an absorption chiller in a combined heat and power plant was studied by Bruno et al (1996) who examined the absorption unit as a steam consumer, where the steam flow rate is a function of the inlet air flow to the gas turbine, ambient air conditions, and refrigeration load. For Bruno et al, (1996) part of the cooling load is the result of cooling the gas turbine inlet air temperature to (15 oC), for a maximum efficiency. He concludes that the absorption chiller is economically viable with the gas turbine when the chiller is used to the cool the inlet air to the turbine to increase generator capacity and also cover additional refrigeration needs.

D.R. Brown et al (1996 ) has conducted a comparative assessment of alternative combustion turbine inlet air cooling system. A software was developed to evaluate nearly 3500 cases of combustion turbines to identify the preferred CTAC technology as a function of application conditions from currently available cooling

technologies and identify application conditions where currently available cooling technologies are not cost-effective, i.e., construction of additional power plant capacity without CTAC would be preferred. The net present value (NPV) and capital cost per incremental kW ($/kW) were calculated for about 3500 cases. This includes 44 industrial turbine simple-cycle cases, 24 aeroderivative turbine simplecycle cases and 288 combined-cycle cases, with each evaluated for five different climatic conditions and with the reference case being either an uncooled or evaporatively cooled plant. CTAC was found to be more cost-effective than simply
building additional uncooled power plant capacity for all application conditions investigated. This included relatively moderate climates and applications where storage was limited (e.g, cooling 24 hours per day).

A semi-closed gas turbine and absorption refrigeration combined cycle was modeled by Joseph J. Boza et al ( 2003 ) using one-dimensional steady state thermodynamics. The cycle performance was then studied by producing

parametric design curves that showed trends and gave an indication of potential optimized system performance. The case of a large engine with a nominal power output of 40 MW was considered. The engine can operate with a thermal efficiency approaching 62% while producing 25% as much 5 rC refrigeration as its nominal power output (roughly 20,000 tons) at 30 rC ambient conditions. The turbine engine can continue to operate in the 60% efficiency range while producing the same amount of 10 rC refrigeration in ambient temperatures as high as 40 rC.

Rivero, Ricardo et al (2003), presented a study in which, an exergo-economic comparison of a combined cycle cogeneration system with an absorption refrigeration turbine inlet air cooling unit has been analyzed. The combined cycle was optimized using a refrigeration unit driven by the exhaust steam from the steam turbine. Two types of absorption refrigeration systems were considered: lithium bromide/water and water/ammonia working mixtures. The exhaust steam of steam turbine is supplied to the absorption refrigeration unit at 120 C to produce chilled water or brine, which is used to cool the turbine inlet air. In the case of a

lithium bromide/water unit the chilled water is at 5 C. In the case of the water/ammonia unit the brine temperature can be as low as -10 C or the ammonia can be evaporated directly in the air cooler. Both types of absorption refrigeration units were considered to use cooling water with a temperature of 32 C, from a cooling tower. The analysis consists in the exergy evaluation of the process streams and considers their economic value, this being known as an exergoeconomic analysis. The comparison between the two combined cycles with the two different absorption refrigeration systems was carried out taking into consideration the operating conditions of each. The results have been compared showing the costs of the streams, identifying those of electricity generation, high pressure steam, gas turbine exhaust gases, stack gases to atmosphere and chilled water or brine as a cold service, all these as a function of air humidity.

Varani et al (2003) presented a study of energetic and exergetic analysis of an single effect Absorption Refrigeration System using the pair Lithium Bromide and Water, based on the available energy coming from the direct burns of the Natural Gas. The energetic analysis has as objective to study the process involved in the several components of the cycle. All simulations were made having as mean entrance parameter the available generator heat, unusually as the observed in the published works that use cooling effect as initial basic. The obtained value of the COP was 0.77. It will have the cooling capacity varying of 14.07 to 49.24kW, that depending of the source of heat used in the generator. The exergetic analysis objectified the quantitative and the qualitative evaluation of the studied system. The component that presented larger irreversibility was the vapor generator, whose relationship with the total irreversibility was around 80%, while the relationship of the other components varies of 0.4 to 8.0%,consequently the smallest exergetic efficiencies was found in the vapor generator, then the optimization of the system should begin at this component.

2-3 Improvements of absorption cooling systems

A small scale experimental single-effect lithium bromide absorption chiller was used for an experiment in the Hot Springs of Sivas, Turkey by Kececiler et al (1999). A computer model was developed based on the experimental data to determine the performance of an absorption chiller. The availability of the geothermal heat source is too low to be used effectively in generating electricity but the hot water source can be used to air condition at 40-50 o F (4-10 o C). The only way they were able to utilize such a low temperature heat source is because of the extremely high warm water flow rate. For example, 100 tons with a temperature drop of 10 o F would require a volumetric flow rate of approximately 300 gpm. A parametric study on the input parameters shows an increase in COP for an increase in generator or evaporator temperatures and a decrease in COP for an inc rease in absorber cooling water temperature.

Mitsuo et al (2003), from the university of Tokyo. Japan presented a study that aimed to obtain the lower evaporator temperature below zero degrees by the waterlithium bromide absorption machine, which is inapt for obtaining the lower temperature in general. In this study pure water as refrigerant in evaporator is mixed with absorbent solution. It has been definitely shown by this study that absorption machine using water-lithium bromide as working fluid can take the lower evaporator temperature below zero degrees, although mixing absorbent solution with refrigerant can cause cycle performance to decrease about 15 % for that in the case without mixing. De Lucas et al (2003), proposed mixtures of lithium bromide and organic salts of sodium and potassium (formate, acetate and lactate) as alternative absorbents with the objective of overpass the limitations of lithium bromide. Lithium bromide presents as inconveniences that can not work below 0C due to crystallization problems and the elevated temperature necessary to concentrate the lithium bromide solution in the generator by boiling. The physical properties of the alternative absorbents has been measured: density, viscosity, enthalpies of dilution,

solubility and liquid-vapour equilibrium data of proposed mixtures. A simulation program has been developed to evaluate the efficiency of the cycle. The physical properties of the mixtures previously measured have been introduced in the simulation program to determine the temperatures and exchanged heats in the different sections of the cooling machine. The main conclusion is that the heating requirements in the generator section can be dramatically decreased using this kind of mixtures, being enough using a waste stream with only 60C to reach the boiling of the diluted absorbent mixture. This fact provides an important economic advantage in absorption cycle.

C. K. Qin et al ( 2006 ) has conducted a study to explore the margin within which gas cooling may defeat its electric counterpart. A hybrid air-conditioning system incorporating engine-driven chiller and desiccant dehumidification was configured out and experimentally tested. Sensible load was undertaken by chilled water from engine-driven chiller, while moisture load was undertaken by recovered waste heat. A test rig was set up to measure the performance of engine -driven chiller and dehumidification capacity. Comparison between theoretically predicted and measured values was presented. The waste heat recovered from engine-cooling system and exhaust was experimentally determined. Also measured was the performance improvement of chiller with higher temperature chilled water being capable of only sensible cooling. Economic benefits of hybrid air-conditioning system over conventional electric chiller were calculated for a reference building in China, on the basis of experimental data. The result reveals more than 30% savings on operation cost can be achieved.

Xiaolin Wang and Hui T. Chua 2009 reviewed more than 100 patents dated mainly from 1985 and later that offer panoply of solutions to better this genre of timely technologies. The surveyed patents were classified into four main categories: (1) absorption system developments, (2) machine components developments, (3) working fluid modifications and additives, and (4) novel applications of absorption cooling systems. Finally they also map out several important directions for future research and development.

The energy and global warming crises have drawn renewed interests to thermally driven cooling systems from the air conditioning and process cooling fraternities. The lithium bromide-water absorption chiller is one of the favorites due to the following specific reasons: (i) it can be thermally driven by gas, solar energy, and geothermal energy as well as waste heat, which help to substantially reduce carbon dioxide emission; (ii) its use of water as a refrigerant; (iii) it is quiet, durable and cheap to maintain, being nearly void of high speed moving parts; (iv) its vacuumed operation renders it amenable to scale up applications. LiBr-H2 O absorption chillers enjoy cooling capacities ranging from kilowatts (kW) to megawatts (mW) which match with small residential to large scale commercial or even industrial cooling needs. However, they currently enjoy only a fraction of the extent of deployment as their vapour compression counterparts. Their major debilitating factors are a low Coefficient of Performance (COP), larger footprint and required headroom, corrosion and crystallization issues and stringent requirements of vacuum leak tightness over its design lifespan. Over the past 30 years, extensive efforts have been devoted to: (i) develop advanced absorption cycles which could work at low heat source temperature or recover more heat to improve system performance; (ii) improve the design of major components such as generator and absorber to enhance their heat and mass transfer efficacy; (iii) avoid crystallization problem; and (iv) develop new and reliable working pairs.

2-4 Modeling and simulation of absorption cooling cycles

Vliet et al (1982) developed a comprehensive computer code to dynamically simulate the water-lithium bromide double-effect vapour absorption cooling cycle and to use the code to investigate the influence of the several design and operational variables on the cycle performance (coefficient of performance and capacity). The heat transfer in the absorber was modelled by a fixed heat transfer coefficient. The results of the work were presented in terms of the effect of each variable on the systems COP.

Takeshita et al (1984) have developed a mathematical computer model of a residential gas-fired absorption heat pump based on R22-DEGDME pair. A prototype absorption heat pump system was designed and tested. The computer simulation and the prototype results were reported to be in a good agreement.

Grossman and Michelson (1985) created a flexible computer simulation program for a vapour absorption system, which makes it possible to simulate various cycle configurations. The user supplies information concerning operating conditions, as well as characteristics of the various sub-units and working fluids of the system. In addition he conveys to the computer an image of the cycle. Based on the state points of the system and the heat quantities at each unit. The Computer code is based on subroutines modelling each component of the absorption system and a main program, which links all the subroutines according to users specifications. The results have been compared with experimental data from tests of LiBr-H2O heat transformer, giving good agreement.

Grossman and Gommed (1987) developed an improved version of the same modular computer code for the simulation of absorption systems, making it possible to predict their performance in varying cycle configurations with different working fluids. The code was employed to investigate the performance of various systems using water-lithium bromide fluid pair. Complete performance maps under varying operating conditions have been generated for systems in single-stage and

10

several double-stage configurations. The code was found to be quite robust with most systems working with non-volatile absorbent. Some problems were encountered concerning fluids with volatile absorbents such as water-ammonia, due to the complex behaviour of those fluids at large concentrations of the volatile component in the vapour phase.

Later, Gommed and Grossman (1990) have modified the same code to simulate the performance of a single-effect and three different double-effect configurations of absorption systems using aqueous lithium bromide. They have found that the parallel double flow configuration yields better results than the series flow one. Recently, ABSIM (1995) (an acronym for ABSorption SIMulation) a modular code developed for simulation of absorption system by Grossman and sponsored by DOE/ORNL. The modular code has been employed successfully to simulate a variety of single-effect, double-effect and dual loop absorption chillers, heat pumps and heat transformers employing the working fluids LiBr-H2 O, H2 O-NH3, LiBr/ZnBr2-CH3 OH, NaOH-H2O and more. Kaushik et al (1985) introduced a computer-aided conceptual thermodynamic design of a two-stage dual fluid absorption cycle for solar refrigeration. This work presented a thermodynamic assessment of a two-stage dual fluid absorption refrigeration system using H2O-LiBr and NH3-H2O as working fluids at the first and the second stage, respectively. The thermodynamic analysis model was based on mass and heat balance equations for two-stage dual fluid absorption system to be carried out. Khatib (1991) developed a theoretical model to describe the steady and dynamic operation of a solar assisted double-effect absorption chiller, with the addition of a storage system, where special attention has been given to the dynamic behaviour of the system components. In addition, an expert system for the absorption air conditioning unit, called the absorption design advisor, has been developed and a case study has been presented.

11

Sabir (1993) reported the development of a computer simulation model of a complete absorption refrigeration system with non-volatile absorbent. The programme calculates the systems operating conditions for given outside and design conditions. The components behaviour has been described in terms of heat and mass transfer effectiveness. The computer simulation model enables the study of the effects of the systems conditions on its performance. Zenouzi et al (1995) reported a dynamic simulation and performance analysis of a double effect lithium bromide-water absorption system. A mathematical model based on mass, species (Lithium bromide), and energy balances for each component of the system is developed to predict the performance of the cycle. The computer code is capable of predicting various heat transfer rates, temperatures, mass flow rates, concentrations, and the storage rates in the refrigerant storage and absorber tanks. The cycle is modeled under two conditions, constant energy input and fluctuating energy input.

Herold (1999) formulated a novel endo-reversible model of an absorption cycle to predict actual cycle performance to within approximately 10% accuracy (for lithium bromide technology). The model also has the feature that it results in an extremely simple result for the coefficient of performance. For example, the model predicts that the COP of a single effect machine should be Te/Th where Te is the evaporator temperature and Th the heat input temperature (absolute temperatures). The model combines extreme simplicity with a very useful upper bound on performance. The accuracy of the model is particularly high considering its simplicity.

A single-effect lithium bromide absorption chiller computer model was validated with experimental data by Homma et al (1994) . The capacity of the absorption chiller is 30 Tons (105.5 kW) and is driven by waste heat in the form of hot water from a gas engine. The absorption model consists of four main components [absorber, generator, evaporator, and condenser]. The model of each component is based on a logmean temperature difference, energy balance, mass balance, salt

12

balance, and an overall heat transfer coefficient. The form of the overall heat transfer coefficient is extra work was 0.15 hp [117 W], but for larger systems this parasitic load might not be an advantage to reducing operating costs.

Glebov et al (2002) presented a tailor-made simulation program Optimization of low temperature driven absorption chiller. This program consists of two main modules: heat transfer coefficients calculations for each unit and mass and heat balances for the whole cycle. In order to complete this simulation program number of experiments and analysis were undertaken utilizing equipment of different scale. A number of tests showed that the district heating temperature has significant effect on the cooling capacity of the chiller. LiBr solution flow rate is also an important factor. It seems that for a certain design configuration and external conditions there is an optimum LiBr solution flow rate in terms of COP value. Test runs with and without subcooler revealed that it was more beneficial for the cooling capacity of the chiller to cool down LiBr solution in an external heat exchanger and utilize the absorber surface for steam absorption. The flow rate effect on the distribution quality and falling film heat transfer coefficient were studied experimentally. Visual observations revealed that when the distribution box is filled with the liquid phase, the heat transfer surface is fully wetted. When the box distributor is partly filled with the liquid phase then dry areas appear on the lamella surface. The overall heat transfer coefficient calculated from experimental measurements appeared to depend on the temperature difference between two phases as well as on the flow rate. In the commercial low temperature driven absorption chiller mean log temperature difference in the absorber and generator is about of 10 oC. Falling film heat transfer coefficients were calculated on the basis of experimental data and attributed to 10 oC mean log T. A correlation between film-side heat transfer coefficient in the form of (Nu) Nusselt number and the flow rate (Reynolds number) was derived. Experimental data about heat transfer in the absorber on a 1,15 MW chiller revealed that dry surface (up to 40 % depends on the flow) occur during operation. The reconstruction of the distributor will lead to the capacity increase.

13

A conform modeling for both ab- and adsorption chillers, has been worked out by Schweigler et al (2003) form the Bavarian Center for Applied Energy Research on the basis of the so-called characteristic equation. By that, means a realistic description of the operational behaviour is obtained taking into account the major irreversibilities of the cycles. A detailed cycle simulation is not necessary. The model only requires specification of operational data of the plants; e.g. solution flow rate or cycle duration, thermal mass of the active sorbent and of passive parts like housing and heat exchangers. Predictions of the model are in good accordance with manufacturer data.

Degang Fu et al (2003), from Shanghai Jiao Tong University, of China presented a numerical study of dynamic of a gas-fired double-effect lithium bromide absorption chiller when it is disturbed by the changes of some running parameters, such as the mass flow rates or temperature of entering cooling or chilled water. A dynamic simulation software has also been developed with FORTRAN and VB languages such that the geometry details, entering cooling and chilled water parameters, gas consumption rate, as well as the solution circulation rate of the chiller can be specified by users. The optimal control strategies, controlling the gas consumption rate and the circulation rate to keep a stable leaving chilled water temperature, have also been presented.

Bruno J.C. et al (1996) have developed a simulation and optimization model for the integration of absorption chillers in cogeneration plants. A simulation model of single effect Absorption Chiller by using Water-Lithium Bromide as working fluid pair has been used to validate the methodology. The final result was a reduction of the energy wastes, and a rise in the overall plant efficiency.

S. Bittanti et al. 2010 have studied the dynamic behavior of absorption chiller with a water/lithium bromide solution. It is described by means of constitutive correlationships and conservation equations. They worked out a detailed model linking the plant dynamic behaviour to the various phenomena and their interactions. The model is implemented using Simulink libraries that interact with

14

several S-functions written in the C programming language. In the study an accurate model was developed, a plant simulator was worked out. Such simulator is most useful to study the dynamics of the plant, so as to asses the applicability and limitations of the refrigeration chiller in air conditioning systems.

15