<%BANNER%>

Performance Enhancement of Semi-Closed Cycle Gas Turbines By Aqueous Glycol Direct Contact Chilling

Permanent Link: http://ufdc.ufl.edu/UFE0044829/00001

Material Information

Title: Performance Enhancement of Semi-Closed Cycle Gas Turbines By Aqueous Glycol Direct Contact Chilling
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Oztekin, Erman K
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: combined -- contact -- cycle -- direct -- ethylene -- exchanger -- gas -- glycol -- heat -- power -- regenerative -- turbine
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The Power, Water Extraction, and Refrigeration (PoWER)cycle is a novel semi-closed cycle gas turbine system integrated with absorption refrigeration. The focus of this study is the feasibility of subfreezing cooling of the gas path using an ethylene glycol-based direct contact heat exchanger (spray cooling). This would allow increased efficiency while avoiding potential turbomachinery damage from ice, but would require a modified absorption refrigeration system to meet both subfreezing and above freezing loads. A model has been developed to explore such a modified design and assess its advantages. In this work, the glycol cooling or the spray cooling will refer to spraying aqueous ethylene glycol into the gas flow inside the direct contact heat exchanger, which is shown spray chamber in the cycle, operating at subfreezing temperatures of water. Two different engine sizes were considered, small (nominally 500kW) and medium (nominally 10 MW). Steady state thermodynamic modeling was applied with state-of-the-art polytropic efficiencies and pressure drops for the turbo-machinery and heat exchangers, approximate equilibrium chemistry for gas properties, and both MEGlobal's and NIST mixture properties for aqueous ethylene glycol solutions. The overall efficiency of the medium size system was estimated to be 47.8% without spray cooling and it rises to approximately 50%with spray cooling, while the overall efficiency of the small system was estimated to be 34.8% without spray cooling and it rises to approximately 37.5% with spray cooling. In addition spray cooling decreases the sensitivity of the system efficiency on input parameters, allowing system optimization based on economic factors.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erman K Oztekin.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Lear, William E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044829:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044829/00001

Material Information

Title: Performance Enhancement of Semi-Closed Cycle Gas Turbines By Aqueous Glycol Direct Contact Chilling
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Oztekin, Erman K
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: combined -- contact -- cycle -- direct -- ethylene -- exchanger -- gas -- glycol -- heat -- power -- regenerative -- turbine
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre: Mechanical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The Power, Water Extraction, and Refrigeration (PoWER)cycle is a novel semi-closed cycle gas turbine system integrated with absorption refrigeration. The focus of this study is the feasibility of subfreezing cooling of the gas path using an ethylene glycol-based direct contact heat exchanger (spray cooling). This would allow increased efficiency while avoiding potential turbomachinery damage from ice, but would require a modified absorption refrigeration system to meet both subfreezing and above freezing loads. A model has been developed to explore such a modified design and assess its advantages. In this work, the glycol cooling or the spray cooling will refer to spraying aqueous ethylene glycol into the gas flow inside the direct contact heat exchanger, which is shown spray chamber in the cycle, operating at subfreezing temperatures of water. Two different engine sizes were considered, small (nominally 500kW) and medium (nominally 10 MW). Steady state thermodynamic modeling was applied with state-of-the-art polytropic efficiencies and pressure drops for the turbo-machinery and heat exchangers, approximate equilibrium chemistry for gas properties, and both MEGlobal's and NIST mixture properties for aqueous ethylene glycol solutions. The overall efficiency of the medium size system was estimated to be 47.8% without spray cooling and it rises to approximately 50%with spray cooling, while the overall efficiency of the small system was estimated to be 34.8% without spray cooling and it rises to approximately 37.5% with spray cooling. In addition spray cooling decreases the sensitivity of the system efficiency on input parameters, allowing system optimization based on economic factors.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erman K Oztekin.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Lear, William E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044829:00001


This item has the following downloads:


Full Text

PAGE 1

1 PERFORMANCE ENHANCEMENT OF SEMI CLOSED CYCLE GAS TURBINES BY AQUEOUS GLYCOL DIRECT CONTACT CHILLING By ERMAN K. OZTEKIN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DE GREE OF MASTER OF SCI E NCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Erman K. Oztekin

PAGE 3

3 To my mom and dad

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my chair Dr. William E. Lear for the assistance and expertise he provided while dea ling with this lengthy study. I really appreciate his efforts to improve my knowledge in this field and his patience in spen ding time with me as my mentor. I also appreciate Dr. Herbert Ingley and Dr. Oscar Crisalle for their willingness to be part of my c ommittee and their efforts in my review process. I especially thank Dr. Ingley for his suggested modification to the cycle which is basis of this study. I thank to my lab mates Sung Joo Hong, Kurt Schulze, and Minki Kim for their friendship and cooperative ness. I want to especially thank Eric W. Lemmon at NIST for his explanations of his mixing model. Finally, I would like to thank to my parents for their priceless love and assistance du ring all of my time in the US. Especially my mom always encourages me i n my professional work life. I really appreciate Yasemin for her support and unique friendship since I have known her.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 NOMENCLATURE ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 2 LITERATURE REVIEW ................................ ................................ .......................... 17 3 BACKGROUND ................................ ................................ ................................ ...... 21 Glycol Properties ................................ ................................ ................................ .... 22 Ethylene Glycol ................................ ................................ ................................ 22 NIST REFPROP ................................ ................................ ............................... 22 Ideal Solution Approximation for Ethylene Glycol Water Mixtures .......................... 23 Partial Pressure Discussion Based on Literature ................................ ............. 23 Molecular Interaction Considerations ................................ ............................... 23 Mixing Enthalpy Considerations ................................ ................................ ....... 24 4 TH ERMODYNAMIC MODEL ................................ ................................ .................. 27 Cycle Configuration ................................ ................................ ................................ 27 Mathematical Model ................................ ................................ ................................ 29 Gas Turbine Cycle ................................ ................................ ............................ 30 Heat Exchangers ................................ ................................ ....................... 30 Compressors ................................ ................................ .............................. 31 Turbine s ................................ ................................ ................................ ..... 31 Pressure Drops ................................ ................................ .......................... 31 Mixing Junctions ................................ ................................ ........................ 32 Vapor Absorption Syst em ................................ ................................ ................. 32 Glycol Cycle ................................ ................................ ................................ ..... 33 Spray Chamber ................................ ................................ .......................... 33 Heat Exchanger ................................ ................................ ......................... 35 Generator EG ................................ ................................ ............................. 35 Pump ................................ ................................ ................................ ......... 36 5 SOLUTION METHOD ................................ ................................ ............................. 39

PAGE 6

6 Design Values ................................ ................................ ................................ ......... 40 Assumptions of the Glycol Cycle ................................ ................................ ............ 41 Validation of the Glycol Code ................................ ................................ .................. 41 6 RESULTS ................................ ................................ ................................ ............... 44 Medium Size Engine ................................ ................................ ............................... 46 Small Engine ................................ ................................ ................................ ........... 49 7 SUMMARY AND CONCLUSIONS ................................ ................................ .......... 62 APPENDIX A COMPUTER CODE FOR ETHYLENE GLYCOL CYCLE ................................ ....... 65 B BASIC CHEMICAL EQUATIONS FOR COMBUSTION ................................ .......... 72 LIST OF REFERENCES ................................ ................................ ............................... 74 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 76

PAGE 7

7 LI ST OF TABLES Table page 3 1 Excess molar enthalpies of different molar aqueous ethylene glycol solutions taken from literature at specific temperature ................................ ...................... 25 5 1 Base values for independent variables ................................ ............................... 43

PAGE 8

8 LIST OF FIGURES Figure page 3 1 Characteristic lines and phases of a binary mixture ................................ ........... 25 3 2 Freezing point of aqueous ethylene glycol solutions at different concentrations ................................ ................................ ................................ .... 26 3 3 Partial pressu re vs. concentration of aqueous ethylene glycol mixtures ............. 26 4 1 Gas flow path of the semi closed cycle ................................ ............................... 37 4 2 Glycol cycle ................................ ................................ ................................ ........ 38 4 3 Control volumes around ethylene glycol cycle ................................ .................... 38 6 1 Thermal efficiency of the system versus recuperator inle t temperature for different solution ratios for medium size engine ................................ .................. 52 6 2 High pressure compressor inlet temperature change with recuperator inlet temperature for medium size engine ................................ ................................ .. 52 6 3 Thermal efficiency of the system versus burner inlet temperature for the medium size engine ................................ ................................ ............................ 53 6 4 Thermal efficiency of the system changing with burner exit temperature for medium size engine ................................ ................................ ............................ 53 6 5 High pressure compressor inlet temperature changing with burner exit temperature for medium size engine ................................ ................................ .. 54 6 6 Thermal efficiency of the system for different LPC ratios ( ) changing with recuperator inlet temperature for medium size engine ................................ ....... 54 6 7 Thermal efficiency of the system versus low pressure compression ratios for medium size engine ................................ ................................ ............................ 55 6 8 Thermal effi ciency of the system versus generator exit temperatures for the medium size engine ................................ ................................ ............................ 55 6 9 High pressure compressor inlet temperature versus generator exit temperature for the medium si ze engine ................................ ............................ 56 6 10 Thermal efficiency of the system versus ambient temperatures for medium size engine ................................ ................................ ................................ ......... 56

PAGE 9

9 6 11 Th ermal efficiency of the system versus recuperator inlet temperature for different solution mass flow rates for small engine ................................ ............. 5 7 6 12 High pressure compressor inlet temperature versus re cuperator inlet temperature for small engine ................................ ................................ .............. 57 6 13 Thermal efficiency of the system versus burner inlet temperature for the small engine ................................ ................................ ................................ ....... 58 6 14 Thermal efficiency of the system versus burner exit temperature for small engine ................................ ................................ ................................ ................. 58 6 15 High pressure compressor inlet temperature versus burner exi t temperature for small engine ................................ ................................ ................................ .. 59 6 16 Thermal efficiency of the system versus recuperator inlet temperature for different LPC ratios ( ) for small engine ................................ .......................... 59 6 17 Thermal system efficiency versus low pressure compression ratio for small engine ................................ ................................ ................................ ................. 60 6 18 Thermal system eff iciency versus generator exit temperature for small engine 60 6 19 High pressure compressor inlet temperature versus generator inlet temperature for small engine ................................ ................................ .............. 61 6 20 Thermal efficiency of the system versus ambient temperature for small engine ................................ ................................ ................................ ................. 61

PAGE 10

10 NOMENCLATURE B Bleeding air mass flow ratio [ ] COP Coefficient of perform ance [ ] f Fuel air ratio [ ] h Specific enthalpy [kJ / kg] Mass flow rate [kg / s] P Pressure [kpa] Heat transfer rate [kW] R Recirculation ratio [ ] T Temperature [K] Power [kW] x Concent ration of ethylene glycol in aqueous solution [ ] Greek Letters Specific heat ratio [ ] Heat exchanger effectiveness [ ] Efficiency [ ] Relative hum idity [ ] Combustor equivalence ratio [ ] Specific volume [m 3 / kg] Specific humidity [ ] Subscripts amb Ambient comp Compressor

PAGE 11

11 EG Ethylene glycol evap Evaporator gen Generator in Inlet max Maximum out Outlet p Polytropic sat Saturation turb Turbine W water Abbreviations AFE Above Freezing Evaporator cm Combustor C1 Low pressure compressor C2 High pressure compressor GENR Generator IC Intercooler RECP Recuperator T1 High press ure turbine T2 Low pressure turbine

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the R equirements for the Degree of Master Of Science PERFORMANCE ENHANCEMENT OF SEMI CLOSED CY CLE GAS TURBINES BY AQUEOUS GLYCOL DIRECT CONTACT CHILLING By Erman K. Oztekin December 2012 Chair: William E. Lear Major: Mechanical Engineering The Power, Water Extraction, and Refrigeration (PoWER) cycle is a novel semi closed cycle gas turbine syst em integrated with absorption refrigeration. The focus of this study is the feasibility of subfreezing cooling of the gas path using an ethylene glycol based direct contact heat exchanger (spray cooling). This would allow increased efficiency while avoidin g potential turbomachinery damage from ice, but would require a modified absorption refrigeration system to meet both subfreezing and above freezing loads. A model has been developed to explore such a modified d esign and assess its advantages In this work the glycol cooling or the spray cooling will refer to spraying aqueous ethyle ne glycol into the gas flow inside the direct contact heat exchanger which is shown spray chamber in the cycle, operating at subfreezing temperatures of water. Two different en gine sizes were considered, small (nominally 500 kW) and medium (nominally 10 MW). Steady state thermodynamic modeling was applied with state of the art polytropic efficiencies and pressure drops for the turbo machinery and heat exchangers, approximate equ ilibrium chemistry for gas properties, and both MEGlobal's and NIST mixture properties for aqueous ethylene glycol solutions. The overall efficiency of the medium size system was estimated to be 47.8% without spray

PAGE 13

13 cooling and it rises to approximately 50% with spray cooling, while the overall efficiency of the small system was estimated to be 34.8% without spray cooling and it rises to approximately 37.5% with spray cooling. In addition spray cooling decreases the sensitivity of the system efficiency on in put parameters, allowing system optimization based on economic factors.

PAGE 14

14 CHAPTER 1 INTRODUCTION The gas turbine is a type of internal combustion engine which has a broad range of industrial usages. Gas turbines are steady flow machines in which gas is gener ally compressed, heated, and expanded for the purpose of generating power or thrust. The basic thermodynamic open gas turbine cycle is the Brayton cycle which is implemented via a gas compressor, a burner, and an expansion turbine. The semi closed cycle is an improvement on the open cycle in which a large fraction of partially expanded gas is recirculated back into the entrance of an intermediate compression stage after cooling and mixing with sufficient fresh air to support the combustion process. Semi clo sed cycle gas turbines are currently of interest due to their potential for improving efficiency and reducing emissions, especially in distributed generation or transportation applications. Generally, combining two or more thermodynamic cycles can raise th e system efficiency by utilizing otherwise wasted heat rejected from the topping cycle. In most combined cycle applications, a gas turbine is the topping cycle and a Rankine system is the bottoming cycle. Instead of using a Rankine cycle to generate power, other cycles can be used as a bottoming cycle, in combination with various modified gas turbine cycles. This thesis deals with a novel combined cycle called t he Power, Water Extraction, and Refrigeration (PoWER) system The PoWER cycle combines a semi clo sed gas turbine system called the High Pressure Regenerative Turbine Engine (HPRTE) with a vapor absorbtion rerigeration bottoming cycle (VARS), to provide power at high efficiency, along with cooling, heat, and fresh water.

PAGE 15

15 Absorbtion refrigeration syste ms can use low grade thermal energy rather than electricity so this makes them attractive to use in conjunction with the gas turbine to utilize waste heat. The VARS only consumes electricity for its pump which requires very low power. Mostly commonly, ammo nia water or lithium bromide water are preferred as the refrigerant/absorbant fluids for vapor absorption systems. For a given VARS design, the cooling capacity of the VARS depends on the available waste heat of the HPRTE cycle. It is well known that lower compressor inlet temperatures lead to higher thermodynamic efficiencies in conventional gas turbine systems. The PoWER cycle is a form of combined cycle in which waste heat from the HPRTE drives the VARS, which in turn provides cooling prior to the main c ompression process. Analogous to providing a low inlet temperature in a conventional gas turbine, this increases the HPRTE efficiency substantially. The inlet of the high pressure compressor can be cooled to a temperature just above the freezing point of w ater by the VARS without forming ice which may damage downstream equipment. The additional VARS capacity is used to provide cooling to external loads. Some applications of the PoWER system do not require external cooling, but would instead benefit from inc reased thermal efficiency. This could be achieved by cooling the HPRTE below freezing, provided that a means can be developed for avoiding ice formation. In this thesis, an innovative approach will be described and analyzed, using a glycol based direct con tact heat exchange process. The development of the PoWER model from previous authors, as well as related technologies, are presented in a literature survey in Chapter 2. The background

PAGE 16

16 knowledge and supporting information about approximations for aqueous g lycol mixture analysis is given in Chapter 3. The PoWER cycle and the glycol cycle overview with the thermodynamics used to model both cycles is provided in Chapter 4. The solution method is described in Chapter 5. The results are shown in Chapter 6 for tw o representative engine sizes. Chapter 7 draws conclusions and makes recommendations for future studies.

PAGE 17

17 CHAPTER 2 LITERATURE REVIEW This section reviews published literature relating to semi closed gas turbine engines, combined HPRTE/VARS cycles, along wi th the dehumidification systems. In addition, published information pertinent to ethylene glycol properties is presented as it applies to the analysis described in this work. Semi closed gas turbine cycles were proposed by Anxionnaz [17 18] as an alternati ve to conventional open cycle gas turbine engines. Semi closed gas turbines have an advantage in size and weight, low fuel consumption over the power range, and significantly reduced airflow requirements. Westinghouse and the U.S. Navy [20] worked on the s emi closed cycle project, named Wolverine. Sulzer Brothers operated power plants with a 5 MW and a 20 MW capacity between 1945 and 1949 for an NOK (Nordostschweizerische Kraftwerke AG) nuclear power plant in Weinfelden in Switzerland. However, they were no t successful due to formation of deposits resulting from burning of crude oil and corrosion of the combustion chamber. The units became dirty and the flow machines suffered excessive wear. Project Wolverine was a development program for a semi closed engin e to be used in submarines, but it was cancelled due to the success of nuclear power plants in that application. Semi closed gas turbine systems can be feasible if a clean fuel such as natural gas is used or if materials with improved corrosion resistance are used. Thus, studies have recently been performed for carbon dioxide gas turbine cycles with internal firing of natural gas and oxygen as an oxidizer [21]. Lear and Laganelli [19] demonstrated a new type of semi closed engine called the High Pressure Re generative Turbine Engine (HPRTE) which includes a low pressure spool that serves to improve power density of the cycle.

PAGE 18

18 They showed that the HPRTE has a weight and volume advantage compared to other recuperated engines, and it has better performance along with low emissions. Much work has been done on semi closed gas turbine systems and some of the different configurations have been developed. Nemec et al. [9] studied the HPRTE cycle with a Ran kine bottoming cycle and they considered several bottoming cyc le fluids. He reported a gross thermal efficiency over 60% using a steam bottoming cycle with superheat. He concluded that it is difficult to match the topping and bottoming cycles near their desired operating points for the component and fluid characteris tics evaluated in his thesis. Boza [8] combined the HPRTE power cycle and a lithium bromide VARS. He simulated both a large engine and a small engine. He reported a thermal efficiency of 62% for his large engine and a thermal efficiency of 43% for his smal l engine without including turbine cooling in his model. In this work, the configuration of the gas turbine portion of the system was developed from Khan et al. [1]. After this initial development, the combination of the HPRTE cycle and a VARS was proposed by. Lear and Sherif in order to improve efficiency and provide cooling to external loads, as well as providing fresh water. Afterwards, experimental efforts on a novel gas turbine engine cycle combined with VARS was conducted by Howell [22 ]. He performed experiments in order to develop and validate design point models for the combined cycle. He tested four engine configurations with vapor absorption system. He resulted that VARS hardware can be successfully integrated with pressurized and semi closed engin e ducting, and driven by the hot recirculation gases. The modeling effort for the design of PoWER cycle was done by Tiffany. He called the cycle with a combination

PAGE 19

19 of power, heat, refrigeration, and fresh wa ter as quad generation cycle [23 ]. He resulted al most constant efficiency across a wide operating range with maximum value of 26.5% integrated with VARS. He expressed that the result of the low efficiency due to the recuperator bypass, the internal leakages, and relatively high pressure drops. To simula te the VARS unit, Second Law efficiency has been used by Choon [2] where he also included the effect of temperature gradients through the thermal reservoirs for his calculation of coefficient of performance. He also simulated a multistage VARS system to ca lculate hourly air conditioning and ice making capacities for a load leveling application. His results show that the daily storage of thermal energy in the form of ice is almost equivalent to the air conditioning supplied directly to the typical building c onsidered. Over the years, turbine inlet temperatures of gas turbines have increased considerably. The development in turbine inlet temperature has led to higher cycle efficiencies. Improved materials and application of superior cooling schemes help withst and higher temperatures. Internal cooling and film cooling increase the allowable gas temperature on the turbine blades. Since 1950, the allowable turbine blade material temperature has advanced approximately 10 C per year. The Advanced Gas Turbine System Research sponsored by the U.S. Department of Energy are planned to achieve turbine inlet temperatures up to 1650 C which can be accomplished by using different cooling schemes for turbine blades to maintain blade temperatures below 704 C [16]. Mitsubishi H eavy industries has achieved the world's highest turbine inlet temperature of 1600 C with achieved power output of about 320 MW gas turbine [24 ].

PAGE 20

20 The cooling air is bled after the final compression stage but it has to be minimized since reduced turbine flo w reduces system performance. The required cooling depends on the temperature after combustion, and it has been estimated to vary from 8% to 14% while the turbine inlet temperature varies from 900 C to 1400C by Massardo et al. [3]. Khan assumed 14% bleed for his model, taken from the compressor exit and rejoining the main stream within the high pressure turbine. Many substances are being used in practical applications of desiccant systems. Solid desiccants are deployed mostly in rotational systems while l iquid desiccants can be sprayed directly into the gas flow path. There is broad usage of chemical solutions in a range of application s from air conditioning to dehumidification systems. Designing systems requir es thermophysical information about working fl uids especially vapor press ures, density, specific heat, etc. for heat transfer purposes. Aqueous solutions are the most common in this kind of systems F or example in absorption dehumidification systems, aqueous solutions of salts or aqueous solutions of organic compounds such as most glycols, triethylene glycol (TEG) and aqueous L iCl are most common [4]. In most applications, moisture is captured inside the conditioner, and the dilute solution is heated inside the generator, releasing the moisture to the scavenger air stream. High affinity and freezing point depression make glycols preferable to mix with water for freeze protection.

PAGE 21

21 CHAPTER 3 BACKGROUND The section deals with ideal approximations used in computations involving ethylene glycol water mixtu res. Ethylene glycol water mixtures and ammonia water mixtures used in absorption refrigeration are both binary solutions with complete solubility in the concentration range of interest. However aqueous glycol solutions can be approximated to be ideal, but ammonia water solutions cannot. The required definitions and the reasons will be discusses throughout this chapter. According to Gibbs Law, when a binary mixture is in its equilibrium state, it has three degrees of freedom. More specifically if three in dependent properties are known it means that other thermodynamic properties of the binary mixture can be found by using relations among properties. Figure 3 1 shows a schematic phase diagram for a binary mixture, relating temperature and composition at a fixed pressure. As see n in the F igure 3 1 t he line of boiling points for different compositions, the bubble line, is not a straight line. Similarly, the initial condensation line, the dew line, is curved. Together the bubble and dew lines define the two p hase region and allow determination of concentration from pressure, temperature, and quality. Additional information about ethylene glycol water solutions and the mixture model in the National Institute of Standards and Technology (NIST) Reference Fluid Th ermodynamic and Transport Properties Database (REFPROP) program will be presented in this chapter. Unlike the straightforward analysis of pure substances, systems consisting multiple components require more complex analysis techniques, since their properti es depend on concentration in addition to two independent state

PAGE 22

22 properties such as temperature and pressure. Ideal solutions are liquid mixtures with thermodynamic property relations analogous to those of a mixture of ideal gases. The ideal solution, which obeys the Raoult's law, has negligible enthalpy of mixing. According to Raoult's law, the vapor pressure of an ideal solution is dependent on the vapor pressure of each chemical component and the mole fraction of the compon ent present in the solution. Thi s simplification allows an alg ebraic phase equilibrium model. Glycol Properties Ethylene Glycol Ethylene glycol is a substance often used to mix with water in order to operate at temperatures below zero Celsius Ethylene glycol itself freezes at about 12 C but when mixed with water, the solution freezing point is further suppresses Figure 3 2 shows the variation of the freezing point with the concentration of ethylene glycol, from MEGlobal's data sheet for aqueous ethylene glycol solutions [13]. The sol ution has a minimum freezing point near a concentration of 60% ethylene glycol. Ethylene glycol, diethylene glycol, and triethylene glycol present no hazard of explosion, polymerization, fire or other industrial risks. Of these, ethylene glycol is the most hazardous to the human body and the environment. Fortunately, ethylene glycol has very low vapor pressure and it cannot escape from the engine system easily. These glycols are all hygroscopic and they are suitable for the current application NIST REFPRO P NIST REFPROP is a key source of information for this analysis, since most of the properties of pure water, ethylene glycol and the ir solution are provided Version 9.0 of REFPROP has the capability to provide thermodynamic data for mixtures of ethylene g lycol and water by using an internal model. Mixing parameters are provided by the

PAGE 23

23 KW0 Kunz and Wagner model for hydrocarbon mixtures. Several model s provided in REFPROP for specific cases but the KW0 and LJ6 models are generalized for hydrocarbons. Lemmon fit the data for ethylene gl ycol and water mixtures and incorporated it into REFPROP [12]. Existing data on ethylene glycol water solutions from various researchers were taken for the model although data available for ethylene glycol water mixtures remai ns limited. So the accuracy of properties for this mixture encounters two important problems. One problem is the limited data to fit the model for the mixture. And the second disadvantage is over fitting of the property model inside REFPROP especially on s ome critical regions of the mixture. Ideal Solution Approximation for Ethylene Glycol Water Mixtures The ideal solution approximation for ethylene glycol water mixtures can be supported for reasons discussed below. Partial Pressure D iscussion Based on Lite rature According to the Trimble and Walter [5] ethylene glycol water mixtures obey R aoult's Law fairly closely for all concentrations. In Figure 3 3, a vapor pressure versus conc entration graph for different ethylene glycol water concentrations at temperat ure 330 K is presented The line in Figure 3 3 is almost a straight line. Molecular Interaction C onsiderations The weak intermolecular bonding of the ethylene glycol and water will be examined in order to shed light on the excess enthalpy of mixing. Water and ethylene glycol molecules form hydrogen O H bonds within their pure liquids Also, water and ethylene glycol molecules are forming hydrogen O H bonds with each other in solution Since they hav e the same bonds with similar and different molecules both water and ethylene glycol molecules have about the same energetics in their pure liquids or in the

PAGE 24

24 mixture. This results in complete miscibility of ethylene glyco l in water. Equally important, also the bond energy is almost same; leading to the conclusion that enthalpy of solution must be quite small. Mixing Enthalpy C onsiderations Ulb ig and Schulz [14] investigated excess molar enthalpies of ethanediol, 1 2 propanediol, 1 2 butanediol plus water solutions for different five temperatures under atmospheric pressur es. For this pressure it was observed that the excess molar enthalpies of ethylene glycol water solutions are almost constant in this temperature range. I n the Table 3 1 excess molar enthalpies at temperature 298.15 K are presented. As seen from th e Table 3 1 the values of excess molar enthalpies of different concentrations were given in Joules per mole, orders of magnitude smaller than the values of heat of solution of common binary mixtures including ammonia water It can be concluded that ethyle ne glycol water solutions hav e negligible excess enthalpies for the current application. Further evidence is that the heats of solution of inorganic compounds in water are three orders of magnitude larger than the heat of solution of organic compounds in w ater [15]. As an organic compound, ethylene glycol has a low enthalpy of mixing compared to inorganic solutes. Worswick and Dunn [6] observed that the excess molar enthalpy of ammonia water solutions is on the order of kilojoules per mole.

PAGE 25

25 Table 3 1. Exce ss molar enthalpies of different molar aqueous ethylene glycol solutions taken from literature at specific temperature Excess molar enthalpy data at T=298.15 K x (mole frac.) H m E (J/mol) x H m E (J/mol) x H m E (J/mol) 0.05 300 0.4 754 0.749 404 0.1 457 0.45 725 0.799 320 0.15 564 0.5 693 0.85 251 0.2 674 0.55 645 0.9 160 0.25 716 0.6 598 0.951 80 0.3 762 0.65 539 0.35 763 0.699 478 Figure 3 1. Characteristic lines and phases of a binary mixture

PAGE 26

26 Figure 3 2. Freezing poi nt of aqueous ethylene glycol solutions at different concentrations Figure 3 3. Partial pressure vs. concentration of aqueous ethylene glycol mixtures

PAGE 27

27 CHAPTER 4 THERMODYNAMIC MODEL Cycle Configuration The PoWER cycle studied in this work is modified r elative to previous studies to allow the VARS output to be utilized completely within the HPRTE, rather than providing cooling to an external load. The modified HPRTE flow path is shown in Figure 4 1. Fresh air at State 1 is taken into the low pressure com pressor (shown in Figure 4 1 labeled C1) and pressurized air discharges into the system at State 2. Then recirculated gas from the recuperator mixes with the inlet air stream adiabatically. The gas s tream is cooled in several steps, from State 2.9 to State 3.2, first by providing high quality waste heat to the VARS (State 2.9 to State 3.0), then rejecting heat to ambient in an intercooler (State 3.0 to State 3.01), then by cooling from the VARS at two temperatures. The air gas mixture is pressurized again i nside the high pressure compressor (HPC, labeled C2). The g as mixture is pre heated in the recuperator (labeled RECP) before the gas enters the combustion chamber. Since the gas temperature is raised before combustion, less fuel is consumed, making the sys tem more efficient than it would be without the recuperator. Combustion (labeled cm) takes place to bring the gas temperature to the desired turbine inlet temperature. The gas is expanded in the high pressure turbine to reach the designated recuperator inl et temperature. This hot flow gives heat to the cold stream of the recuperator. After the recuperator, the flow splits low pressure turbine (LPT, labeled T2) utilizes exhaust gases to drive the low pressure compressor. The bleed percentage was taken to be 14% of the HPC gas flow (the same as in Khan's model) and a high percentage was chosen consistent with the high values of the

PAGE 28

28 turbine inlet temperature. Cooling flow is bled at the combustor inlet. The code is also capable of bleeding from the exit of the high pressure compressor. To penalize the cycle in a conservative way, the bleed flow rejoins the main stream at the exit of the turbine. Expanded gas inside the turbine and the bleeding air reaches the recuperator inlet temperature after mixing adiabatic ally. This approach allows for cooling all of the HPT stages. The model is based on the assumption that uncooled turbine blades have similar capability as the uncooled recuperator material in terms of the maximum temperature allowed. In this new model, t wo usable waste heat exchangers are included (GENR_EG, both shown in Figure 4 1, and 4 2, and GENR_VARS) since heat is required for both the VARS generator and glycol cycle water removal process. Heat is first provided to the glycol cycle since the bubble point temperature of the glycol water solution ranges from 100 C to 198C, requiring the highest temperature waste heat. Then heat is provided for the VARS in its generator. The typical temperature of the State point 2.9 is around 450 C. The location of the generator was selected in order to match the gas and generator temperatures with minimum irreversibility. The HPRTE cycle is combined with the VARS along with a glycol cycle to allow sub freezing temperatures at State 3.2 without ice formation. For gen erality, a Second Law efficiency approach is used to model the VARS refrigeration cycle. On the other hand, the aqueous glycol cycle configuration is fully specified as shown in Figure 4 2. The binary solution is sprayed into the gas flow in the spray cham ber (labeled SP CHAM in Figure 4 1 and Figure 4 2). Cooled and dried gas leaves the spray chamber, typically at subfreezing temperatures, and enters the high pressure compressor. The

PAGE 29

29 diluted glycol solution enters the regenerative heat exchanger of the gly col cycle (labeled HEX) where it is heated. The solution temperature rises higher than the bubble point, allowing water vapor to be extracted in the separator (labeled SEPARATOR). Hot solution returns to the HEX to provide heat to the incoming cold solutio n. After cooling, the flow leaves the heat exchanger and is cooled further in the evaporator by the VARS. Then solution is pumped back into the spray chamber, completing the cycle. Both evaporators that follow the generator and interact with the HPRTE cycl e have the desired effect of reducing the temperature of the hot gas stream entering the high pressure compressor. The intercooler of the semi closed cycle provides natural cooling down to near ambient temperatures. The AFE which is a part of the VARS cool s the gas path as much as possible while remaining just above the freezing temperature of water while most of the water vapor is condensing. Sensible heat of the gas flow and latent heat of the water vapor determines the cooling capacity of this unit, prov ided by the VARS. The temperature of the exit of the AFE is limited so that ice does not form. The spray chamber dries the gas flow further and cools it as much as possible depending on the available waste heat driving the VARS. The low pressure compressio n ratio affects the generator inlet temperature which is higher than atmospheric pressure. This leads to more compact heat exchangers in the semi closed cycle than typical open cycles. Mathematical Model The HPRTE and glycol cycles were modeled with tradit ional zero dimensional steady state thermodynamics. Cycle configurations, approximations, and properties models were presented in the beginning of this section and previous chapters. In order to explore the advantage of including a direct contact heat exch anger via spraying aqueous glycol, the First Law efficiency of the system will be obtained by constructing

PAGE 30

30 thermodynamic models for every component inside each cycle, coupled according the constraints of the cycle configuration. The model was constructed using polytropic efficiencies for compressors and turbines, pressure drops for heat exchangers, energy and mass conservation equations, the effectiveness definition, second law efficiency for the refrigeration cycle, the curvefit for enthalpy calculations, and property models used to get the thermodynamic properties of the glycol solution. In addition, the ambient temperature was specified, as well as the HPT inlet temperature and the recuperator hot side inlet temperature, both materials limited. For each state of the gas flow path, the curvefit was used to calculate enthalpies as a function of temperature and local equivalence ratio. Solution mass flow rates were obtained by applying conservation of mass over multiple control volumes on the glycol cycle. S ince the gas can carry a significant amount of water, psychometric relations were used to find the mass flow rate of the water inside the gas. The properties of aqueous ethylene glycol were taken from NIST REFPROP and MEGlobal's ethylene glycol property eq uations. A second law approach was used for refrigeration systems. The most efficient heat engine that operates between hot and cold reservoirs is the reversible device. Any refrigerator that operates within the same reservoirs can be related to the revers ible (designated here as Carnot) device with the second law efficiency parameter. Gas Turbine Cycle Heat Exchangers The effectiveness method is convenient for relating heat exchanger inlet temperatures to the corresponding exit temperatures, based on the d imensionless

PAGE 31

31 parameter effectiveness. Heat transfer effectiveness is defined as the ratio of actual heat transfer rate to maximum possible heat transfer rate. (4 1) Compressors Polytropic efficiency was used to model compression stages since actual compressors do not follow the isentropic process. The polytropic efficiency relates the compressor pressure ratio to the temperature ratio and the average specific heat ratio, as (4 2) In equation 4 2, P OUT and P IN are outlet and inlet pressu res respectively. T symbolizes p Turbines Polytropic efficiency is again input for turbines to relate outlet and inlet temperatures to outlet an d inlet pressures, given by (4 3) Pressure Drops Relative pressure drops for the heat exchangers and the burner were specified as ratios of pressure drop across the component to the absolute inlet pressure. Hence output absolute pressure is calculated from the following equation

PAGE 32

32 (4 4) Mixing Junctions At the mixing junctions both conservation of mass and energy were applied. Mass flow ratios were used instead of actual values of the mass flow rates sin ce the mass flow rates of every state were normalized to the mass flow of incoming fresh air. (4 5) (4 6) Vapor Absorption System In this work, the second law approach with the second law efficiency was used to model the VARS. The conventional coefficient of performance parameter is defined as (4 7 ) is the cooling provided by the evaporator and is the corresponding heati ng rate that has to be provided to the VARS to accomplish this cooling. Here, the second law efficiency is defined by rationing the actual coefficient of performance to the coefficient of performance of the Carnot cycle as follows: (4 8 ) The coefficient of performance of the Carnot cycle can be calculated by considering the temperature gradients of the hot and cold reservoirs, which here are both proces ses within the HPRTE gas path.

PAGE 33

33 (4 9 ) Where T amb is the ambient temperature, T gen,in and T gen,out are generator gas path inlet and outlet temperatures, and T evap,in and T evap,out are evaporator gas path inlet and outlet temperatures, respectively. Glycol Cycle After being cooled by the VARS, chilled gas co ming from the AFE enters into the spray chamber to be cooled below freezing and to be further dehumidified The gas temperature at the entrance of the spray chamber was designed to be 3 C. At that temperature the gas can still carry a small amount of water vapor. Gas leaving the spray chamber was assumed to be dry and the small escape of the ethylene glycol was ignored. The flow rate of solution sprayed into the gas in the spray chamber was determined based on the balance between VARS capacity and the sensi ble and latent heat transfer from the gas. The solution leaves the spray chamber slightly more dilute and continues through the cycle. Control volumes selected on the system is shown in Figure 4 3. Spray Chamber This component is the most complex one in th e glycol loop since it involves simultaneous heat and mass transfer. Inside the spray chamber, two flows are mixing and water vapor in the gas condenses. A simplified energy balance was applied to this unit with good resulting accuracy.

PAGE 34

34 Considering the spr ay chamber as a control volume, continuity can be written by considering that all species are conserved, and assuming that water is the only species that changes phase. It was assumed that no ethylene glycol leaves the unit inside the gas as discussed prev iously due to the low vapor pressure of ethylene glycol. For a flow containing k different non reacting species, the total mass flow rate is (4 10 ) The existing species entering and leaving the spray chamber are the dry gas mixt ure, ethylene glycol, and water. State 9EG and State 1EG have liquid phases of ethylene glycol and water. State 3.02 has gas phases of the dry gas mixture and water. And finally State 3.2 has only the dry gas phase per the assumption of complete water remo val. The flow rate of water vapor entering the spray chamber is a function of pressure and temperature of the gas flow. Saturated air can carry specific amount of water vapor at given temperature. The humidity ratio can be calculated from relative humidity definition: (4 11 ) (4 12 ) (4 13 ) Here, the gas molecular weight is assumed to be the same as air, so that the constant 0.622 is unchanged. This assumption is appropriate f or combustion gases of typical hydrocarbon fuels. The total energy balance around the spray chamber can be written as

PAGE 35

35 (4 14 ) Enthalpy of the gas flow was calculated as a function of temperature and pressure whereas enthalpy of t he solution was calculated as a function of temperature, pressure, and concentration from MEGlobal's equations [13], (4 15 ) (4 16 ) Heat Exchanger The heat exchanger transfers the glycol solution waste heat to preheat the incoming cold flow from State 1EG The solution coming from the separator enters the heat exchanger and transfers its heat to the solution coming from the spray chamber to decrease the heat that has to be added inside the generator fro m an external source. The effectiveness definition is used to determine exit temperatures. The temperature of State 4EG has to be guessed at the first iteration to apply energy balance on heat exchanger. The effectiveness definition is related to temperatu re change for both streams, for sensible heat transfer only, as (4 17 ) Generator EG After the pressure is decreased across the valve at State 2.1EG, heat is provided inside the generator to bring the solution to the desired temp erature for rejecting water vapor at the same rate that it was absorbed. Since this temperature depends on three

PAGE 36

36 independent variables, all of them must be determined. Concentration inside the generator is known since mass flow rates are known. Pressure is also known since it was an input parameter. The value of the quality of the vapor has to be determined as the last parameter. The quality can be found from the fact that same amount of water coming inside the system has to be rejected: (4 18 ) (4 19 ) Pump The pump provides the pressure increase for overcoming the pressure drops in the spray chamber and other glycol loop components. Isentropic pump work can be calculated from pressure differe nce across it and it can be related to the actual work by using the adiabatic efficiency parameter: (4 20 ) (4 21 )

PAGE 37

37 Figure 4 1. Gas flow path of the semi closed cycle

PAGE 38

38 Figure 4 2. Glycol cycle Figure 4 3 Control volu mes around ethylene glycol cycle

PAGE 39

39 CHAPTER 5 SOLUTION METHOD Two computer codes were written to simulate the performance of the combined cycle, implementing the thermodynamic model described in Chapter 4. The first code was written in Excel to simulate gas flow path of the PoWER cycle, and the second code was written in MATLAB to simulate the ethylene glycol cycle. The PoWER cycle code is capable of calculating net power output, water extraction rate, and system efficiency as a function of the i nput parameters (low pressure compression ratio, burner exit temperature, recuperator inlet temperature, ambient temperature, and recirculation ratio). Additionally, the glycol code is capable of determining the heat that is required to reject sufficient w ater vapor from the separator that the cycle remains in water balance. It also calculates the heat that must be taken from the ethylene glycol cycle as a function of input parameters (solution to air ratio inside the spray chamber, concentration, pressure of the heater, and ambient temperature). The pressures of the state points of the gas turbine cycle were calculated first. Pressure losses across components, burner exit temperature, and recuperator inlet temperature are defined. Once these are known, the values of pressures for all state points can be calculated. Pressures at states 2.9, 2.91, 3.0, 3.01, 3.02, 10, 7.2, and 7.1 can be calculated from pressure losses inside components using Equation 4 4, and pressure drop across the high pressure turbine can be calculated from inlet and outlet temperatures using Equation 4 3. With all pressures known, temperatures of the states were found using effectiveness, polytropic turbo machinery relations, and definitions. Generator exit

PAGE 40

40 temperature and AFE exit tempe rature are defined based on current VARS technology and the freezing point of the water vapor that exists inside the gas. AFE exit temperature was assumed to be 3 C, and all of the water was assumed to be started extraction at the inlet of this component. Temperatures of the states around the recuperator were calculated by using effectiveness and energy balance equations. Temperature of the exit of the compressors and turbines were calculated by polytropic relations by using the known pressure changes. Ent halpies of each state were calculated by using the curvefit as a function of equivalence ratios at the corresponding state points. Exit temperature of the spray chamber depends on the useable waste heat of the generator, which is determined by the glycol c ycle. Then the solution iterates the temperature at the state point 2.9. A balance of heat transfer rates between the semi closed cycle, VARS, and glycol cycle deter mine the value of the high pressure compressor inlet temperature The generator provides he at to the glycol cycle that brings the solution to the desired concentration Then t he rest of the heat is used for cooling the AFE and the spray chamber. The coefficient of performance of the VARS was calculated from Second Law analysis from Equations 4 7 4 8, 4 9. The calculation has converged when useable wa s te heat for cooling of the spray chamber, which is calculated from the gas cycle, equals the cooling heat required for the glycol cycle in the same spray chamber. Design Values It must be ensured that the temperature of the coldest state of the ethylene glycol cycle is above the freezing point of the corresponding solution that has the same concentration. The mass flow rate of the glycol solution must be adjusted to control the

PAGE 41

41 temperature at the e xit of the spray chamber to improve the efficiency of the system. Pressures in the glycol system must be defined before executing the program. Hence, the input pressure is an independent variable to determine other properties of the mixture of each state p oint. Table 5 1 shows base case values for input parameters for the small and medium size engine. Assumptions of the Glycol Cycle The following assumptions were made during the analysis: Since p ressure losses inside the components were neglec ted, the cycle operates at only two different pressures. The first is the pressure of the gas coming from the combined cycle and the second is the generator pressure when water is separated from solution. Exit streams of the spray chamber were assumed to be in equilibr ium. Heat of solution of the aqueous ethylene glycol solution was neglected. The temperature change due to heat releases during the mixing process was shown to be sufficiently low. It was assumed that the gas flow enters spray chamber containing small amou nt of water vapor and leaves the chamber as dried flow. It was assumed that the latent heat of the water vapor joining the solution is much higher than the heat of mixing. Validation of the Glycol Code Since the ethylene glycol system has not been analyzed before investigation of the model accuracy is required The f ollowing cases were examined for different variables by applying an energy balance to check the error of the glycol cycle. When applying the energy balance it should be noted that wa ter vapor enters the system as satu rated vapor and leaves as su perheated vapor. E nergy balances were applied to the following:

PAGE 42

42 The whole system has heat input from the generator of the ethylene glycol cycle and heat output from the evaporator and ambient heat exchan ger M ass flow enters from the gas flow path and exits to the same gas flow path and the ambient Half of the system consists of a separator, generator, valve and heat exchanger. The rest of the system consists of a spray chamber, pump, evaporator, and a mbient heat exchanger. The whole system without cooling in the spray chamber and without water vapor input. F or all cases relative errors were found low compared with the heat input to the glycol cycle. In general, the errors normalized with generator hea t flow rate are lower than 1%, and the errors of the control volume on the whole system are fluctuating around 1%.

PAGE 43

43 Table 5 1. Base values for independent variables Medium size Engine Small Engine Turbine Inlet Temperature [K] 1722 1500 Recuperator Inl et Temperature (Hot Side) [K] 1000 1000 Ambient Temperature [ C] 30 30 Generator Exit Temperature [ C] 100 100 Above Freezing Evaporator Temperature [ C] 3 3 LPC Pressure Ratio 3.5 3.5 Combustor Equivalence Ratio 0.9 0.9 Ratio of mass flow rate of w ater extracted to mass flow rate of fuel 1.51 1.51 Turbomachinery polytropic efficiencies 0.9 0.84 Effectiveness of recuperator 0.9 0.9 Effectiveness of intercooler 0.9 0.9 Pressure drop in recuperator hot side 3% 3.5% Pressure drop in recuperator co ld side 3% 3.5% Pressure drop in combustor 5% 5% Pressure drop in generator 3% 3.5% Pressure drop in above freezing evaporator 3% 3.5% Pressure drop in intercooler 3% 3.5% Bleed fraction 0.14 0.14 Ratio of mass flow rate of solution to mass flow rat e of air inside spray chamber 0.3 0.3 Concentration of ethylene glycol in aqueous solution 60% 60% Low pressure value of glycol loop [kpa] 200 200 Effectiveness of heat exchanger in glycol loop 0.9 0.9 Isentropic efficiency of solution pump 0.8 0.8

PAGE 44

44 C HAPTER 6 RESULTS In this study, two different cases were considered for the application of glycol cooling in the PoWER cycle. The first case is a small engine with a nominal power output of 500 kW, and the second case is a medium size engine with a nominal power output of 10 MW. These cases were chosen for study because the PoWER cycle is expected to be deployed in a distributed generation mode, where large engines are not viable. Thermal efficiency of the combined system is the figure of merit chosen to un derstand the effect of the spray cooling usage. Efficiency of conventional gas turbine engines is known to increase with decreasing temperature at the inlet of the compressor. The primary effect of subfreezing temperatures in the PoWER cycle at the HPC inl et is expected to be analogous to low ambient temperatures for conventional engines. Hence the effect of the glycol cycle on lowering the HPC inlet temperature is also of interest. Thermal efficiency of the system is defined as the net power output of the system divided by the heat input to the system based on the lower heating value of the fuel: (6 1) The goal here is to quantify the effect of the input parameters as well as the addition of the glycol cooling on the system on the system efficie ncy. The effective efficiency is defined so that the excess refrigeration is taken into account by including in the power output the power that would be required to produce the VARS external cooling by

PAGE 45

45 means of a vapor compression system. In this work effective efficiency was not considered. Boza showed that effective efficiency and the ordinary thermal efficiency differ by 2 3%, similar to the difference between thermal efficiencies of spray cooled and non spray cooled combined cycles. In this work, only the thermal efficiencies are compared for the above freezing and spray cooled cases. Note that the comparison therefore ignores the benefit of the above freezing case producing external cooling. In applications th at require cooling, this must be considered. Some material temperature limitations affect the efficiency of the system, as for the case of traditional gas turbines. Recuperator inlet temperature T7.1 and burner exit temperature T6 are limited based on mate rial considerations. In addition, T6 can be higher than material limits due to blade cooling technologies. Another important limitation is the freezing point of the solution. Since the system operates at sub freezing temperatures, a glycol solution freezin g limitation must be considered. Solution mass flow rate relative to the spray chamber gas flow rate has a large effect on the minimum solution temperature. The relative mass flow rates of the solution and gas are defined as (6 2 ) As will be shown, the value of affects the system efficiency, with a lower value of yielding higher efficiencies. But as is decreased, the glycol solution approaches its freez ing point, so the concentration of the ethylene glycol was taken based on freezing point considerations. Some parameters, for example, burner exit temperature T6, recuperator inlet temperature T7.1, and compressor pressure ratios have the highest impact on the

PAGE 46

46 system efficiency because they can both affect the gas turbine cycle efficiency and available waste heat for cooling. On the other hand, some parameters increase gas turbine cycle efficiency but decrease useable available waste heat or vice versa. Bur ner inlet temperature also has an important effect on the system since the entropy generation depends on it, according to (6 3) Entropy change for a system depends most sensitively on the burner inlet temperature, since that i s the minimum temperature at which heat is added to the cycle from an external source. The COP of the refrigeration cycle is sensitive to the ambient temperature, inlet temperat ures of the hot reservoirs, and output temperatures of the reservoirs. As ment ioned before, the temperature gradients of the reservoirs were also taken into consideration. COP has an effect on the available waste heat for the refrigeration. For both medium size and small engines, some graphs show the PoWER cycle with and without gly col cooling. When spray cooling was not applied on the system, all waste heat was considered given to the ambient. Instead of using waste heat for the air conditioning, this heat was used for cooling the spray chamber. The increase in efficiency is illustr ated in the figures at the end of the chapter. Medium Size Engine The first half of the graphs were prepared for the medium size engine case. Important input parameters were chosen for the investigation system efficiency. In Figures 6 1 and 6 2 the recup erator inlet temperature will chance. Burner inlet

PAGE 47

47 temperature will be shown in Figure 6 3 to understand the effect of entropy generation. Figures 6 4 and 6 5 are showing the effect of burner exit temperature. In Figures 6 6 and 6 7 the effect of low pre ssure compression ratio will be shown. The effect of generator exit temperature will be presented in Figures 6 8 and 6 9. At the end the effect of ambient temperature will be shown. Figure 6 1 shows the dependence of the system efficiency on the recupera tor temperature, for several values of the solution gas flow ratio From Figure 6 1, it is seen that for small values of the system has a better efficiency. The selection of de pends on the permitted recuperator temperatures on the system. As discussed before, blade coolant temperature and the expanded gas temperature will be the same with the recuperator inlet temperature at the end of the blade passages after mixing. So both ma terial limitations for the turbine blades and recuperator materials play an important role in that case. When the value of rises, the system can run at lower recuperator inlet temperatures. However, the gain in efficiency is less with glycol cooling when compared to that of a system with lower values of With =0.3, the system can work on a broad range of recuperator inlet temperatures, and the increase in system efficiency is ar ound 1.6% compared to a PoWER cycle with above freezing HPC inlet temperatures. Selecting =0.19 is better since the system efficiency is higher at that value. For lower values of it seems that the system does not gain much more in efficiency. The system efficiency shows peak around the recuperator inlet temperature of 1160 K. Around the peak, glycol cooling does not bring much efficiency gain as it does at lower recuperator inlet temperatures. This peak e fficiency also involves a cost penalty for high temperature materials and may not be feasible. For this case,

PAGE 48

48 optimization is needed to pick an affordable and an efficient glycol cooled system. To summarize, when the recuperator inlet temperature rises, sy stem efficiency also rises toward a maximum. After reaching a peak, efficiency gets lower for fixed turbine inlet temperatures. If the system efficiency becomes higher, the glycol cooling brings less efficiency gain. But glycol cooling provides a flatter e fficiency peak than the standard above freezing case. Figure 6 2 and Figure 6 3 are supplementary graphs for the changing recuperator inlet temperature case. Figure 6 2 shows that a decrease in recuperator inlet temperature decreases the high pressure comp ressor inlet temperature. This happens because when recuperator inlet temperature lowers the system has more useable waste heat to cool down the flow temperature. Additionally, Fig ure 6 3 shows that higher burner inlet temperature brings higher efficienci es to the system w hich is also better in terms of entropy generation. As seen from Equation 6 3 as temperature of the closed system rises the entropy generation decreases. Figure 6 4 shows the effect of the burner exit temperature on the system efficiency Higher values of burner exit temperature are generally better in terms of efficiency. Since both gas turbine cycle efficiency and useable waste heat increase with this value, this temperature is only limited by the material properties. Figure 6 5 validat es the expectation that when burner exit temperature becomes higher, high pressure compressor inlet temperature becomes lower. Figure 6 6 shows an interesting result for the medium size engine case. It shows how system efficiency changes with different low pressure compression ratios. Due to the fact that higher pressure compression ratios bring the VARS system more heat, the

PAGE 49

49 solution freezes for higher recuperator inlet temperatures. So it is not possible to work at recuperator temperatures as high as 100 0 K for small values of The interesting result occurs at around 1000 K where the system efficiency becomes insensitive to changes in the low pressure compression ratio. In fact, for low pressure compression ratios higher than 3. 5, the cycle efficiency decreases. However, the available waste heat increases. So for that case the effects offset each other. It seems that values of low pressure compression ratio near 6 have higher efficiency than the low pressure compression ratio of 3.5 for the small region. Figure 6 7 shows this effect. When low pressure compression ratio increases, the PoWER cycle reaches a peak efficiency and drops to a lower value. But when glycol cooling is applied, even if the PoWER cycle efficiency starts dropp ing from its peak, glycol cooled system efficiency still rises and reaches its own peak value. Then the efficiency of the glycol cooled system drops less steeply, the broad peak described earlier. Figure 6 8 and Figure 6 9 show the generator exit temperatu re effect on the system. It only affects available waste heat, and when it decreases, the system has more usable waste heat. This raises the system efficiency slightly and lowers the high pressure compressor inlet temperature. Figure 6 10 shows the ambient temperature change versus system efficiency. Ambient temperature affects both PoWER cycle and useable waste heat since it also changes the COP of the VARS system. As expected, system efficiency becomes lower when the ambient temperature increases. Small E ngine The same approach used in the medium size engine case was repeated for the small engine case Same trend will be used as in medium size case to present results.

PAGE 50

50 In other words the order of the interested variables will be same. Figure 6 11 presents b ehavior similar to the medium size engine case. Since in this case T6 is lower, efficiency drops significantly after the value of 1111 K of the recuperator inlet temperature. The peak of the system efficiency for a small engine as a function of recuperator inlet temperature is close to the allowable temperatures of the materials. Glycol cooling brings more than a 2 % efficiency gain to the system for recuperator inlet temperatures around 1000 K. So glycol cooling produces a greater benefit when used in smal ler engines. Again, lowering increases efficiency. A value of =0.19 was selected as the best case for the small engine since it is more efficient and the solution does not freeze within the operating tem perature range. If =0.3, the operating temperature range is broader, and the solution can be used without freezing with a greater margin. Figure 6 12 and Figure 6 13 show that, as recuperator inlet temperature decreases, high pre ssure compressor inlet temperature also decreases. Also, for higher burner inlet temperatures, the system is more efficient. The small engine has an efficiency of around 0.39 with =0.19, which is lower than that of the medium size engine as expected. The effect of burner exit temperature can be seen in Figure 6 14. For the small engine as for the medium size engine, increasing burner exit temperature generally has a positive effect on system efficiency. Figure 6 16 shows the expect ed effect on the system. When pressure ratio of the low pressure spool increases, system efficiency tends to drop even when the system has more available waste heat.

PAGE 51

5 1 In Figure 6 20 it is seen that system efficiency decreases for higher ambient temperature s as expected.

PAGE 52

52 Figure 6 1. Thermal efficiency of the system versus recuperator inlet temperature for different solution ratios for medium size engine Figure 6 2. High pressure compressor inlet temperature change with recuperator inlet temperature for medium size engine

PAGE 53

53 Figure 6 3. Thermal efficiency of the system versus burner inlet temperature for the medium size engine Figure 6 4. Thermal efficiency of the system changing with burner exit temperature for medium size engine

PAGE 54

54 Figure 6 5. High pre ssure compressor inlet temperature changing with burner exit temperature for medium size engine Figure 6 6. Thermal efficiency of the system for different LPC ratios ( ) changing with recuperator inlet temperature for medium siz e engine

PAGE 55

55 Figure 6 7. Thermal efficiency of the system versus low pressure compression ratios for medium size engine Figure 6 8. Thermal efficiency of the system versus generator exit temperatures for the medium size engine

PAGE 56

56 Figure 6 9. High pressure c ompressor inlet temperature versus generator exit temperature for the medium size engine Figure 6 10. Thermal efficiency of the system versus ambient temperatures for medium size engine

PAGE 57

57 Figure 6 11. Thermal efficiency of the system versus recuperator i nlet temperature for different solution mass flow rates for small engine F igure 6 12. High pressure compressor inlet temperature versus recuperator inlet temperature for small engine

PAGE 58

58 Figure 6 13. Thermal efficiency of the system versus burner inlet tem perature for the small engine Figure 6 14. Thermal efficiency of the system versus burner exit temperature for small engine

PAGE 59

59 Figure 6 15. High pressure compressor inlet temperature versus burner exit temperature for small engine Figure 6 16. Thermal e fficiency of the system versus recuperator inlet temperature for different LPC ratios ( ) for small engine

PAGE 60

60 Figure 6 17. Thermal system efficiency versus low pressure compression ratio for small engine Figure 6 18. Thermal syst em efficiency versus generator exit temperature for small engine

PAGE 61

61 Figure 6 19. High pressure compressor inlet temperature versus generator inlet temperature for small engine Figure 6 20. Thermal efficiency of the system versus ambient temperature for sm all engine

PAGE 62

62 CHAPTER 7 SUMMARY AND CONCLUSIONS A PoWER variant consisting of the HPRTE semi closed cycle gas turbine, VARS, and glycol cycle was modeled by using zero dimensional steady state thermodynamics. The VARS was modeled using a second law efficienc y approach. Two cases were considered: a medium size and a small engine. In contrast to previous studies, all of the refrigeration capacity was used to cool the high pressure compressor inlet to below ambient temperature. Cooling of the gas was provided b y the VARS system both directly and indirectly. Final drying of the gas flow and depression of the temperature below freezing were accomplished by a direct contact spray chamber. The VARS cycle was operated on the usable waste heat of the semi closed gas t urbine cycle. A thermodynamic model was developed for the complete system, and parametric results were generated. The following are conclusions drawn from this analysis: The medium size engine with 14% blade cooling has a predicted efficiency of 47.8% with out spr ay cooling. Spray cooling raises system efficiency up to 50% The small engine with 14% blade cooling of the compressed air has an efficiency of 34.8% without spr ay cooling. Spray cooling raises sy stem efficiency up to 37.5%. S pray cooling results i n a greater system efficiency increase for the small engine than for the medium size engine. Spray cooling provides a flatter maximum efficiency peak for both sizes. For different ambient conditions the spray cooled system efficiency is insensitive to amb ient temperature for both small and medium size engines. So the system can be run under these conditions with stable performan ce even while the temperature during a particular day changes Increasing temperatur e at the exit of the burner always has a posit ive effect on system efficiency whereas recuperator inlet temperature exhibits an optimal value. Spray cooling becomes useless after that peak value since the system does not have much available waste heat.

PAGE 63

63 The temperature at the inlet of the high pressur e compressor drops to around 13 C for both small and medium size engines for given base case conditions. This temperature is very sensitive to input parameters which are recuperator inlet temperature, burner exit temperature, and generator exit tem perature The solution mass flow rate must be determined by considering both the cooling effect of the spray chamber and the freezing point of the solution. The higher mass flow rates decrease the efficiency gain of the system, however, the solution does not freeze as easily. Mass flow rate must be selected based on the design point of the system. If the design point promises more available waste heat, then higher solution mass flow rates can be used. Future Work : In this study the system efficiency of the spray cooling with the PoWER cycle was compared to the cycle without the spray cooling. This comparison could be extended by applying an optimization method for the various input parameters. For example, the LSGRG2 code has been applied by Khan et al.[7] and a similar approach applied to the spray cooling cycle should produce a global maximum in the cycle efficiency. A simplified model was used for the spray chamber, so accuracy would be improved by increasing the sophistication of this component model. T he more complex model can consider both heat transfer and mass transfer effectivenesses. Additionally, the mixing enthalpy can be included into the analysis; however the sensitivity of the system outputs to these modifications should be estimated before em barking on this effort. The environmental effect of ethylene glycol is an another important issue to consider in the overall analysis. In reality, a small amount of ethylene glycol escapes from the system due to incomplete capture of spray droplets and inc omplete combustion of the droplets in the burner. Because of this, the cost and environmental impact must be considered before implementation.

PAGE 64

64 The required heat for the generator of the glycol cycle is dependent on its pressure. So the effect of the genera tor pressure, which is lowest pressure in the cycle, should be part of the optimization procedure. It is expected that lower generator pressure would decrease the heat input to the generator from the gas path. Also, lower pressure will yield a lower bubble point for the solution which allows use of lower temperature waste heat, possibly lower than that usable by the VARS. The tradeoff is the increased solution pump power.

PAGE 65

65 APPENDIX A COMPUTER CODE FOR ETHYLENE GLYCOL CYCLE Main Code % PROGRAM of ethylene g lycol cooling system % Written by Erman K. OZTEKIN ,Energy and Gas Dynamics lab. % University Of Florida Mechanical Engineering Dept. clear all close all clc global m_8_EG m_8_w deltaT_y T_def T_amb eta ksi global phi v_1 v_11 v_10 v_9 z_1 z_11 z_10 z_9 Q_hex global W_hex W_heater m_3_vapor_EG m_5_EG g global error difference global P_1 P_2 P_3 P_4 P_5 P_6 P_7 P_8 P_9 P_10 P_11 %State point 10 corresponds to 3.02 in combined cycle and state %point 11 corresponds to 3.2 in combined cycle. % --------------------------------------------------------Recirc=xlsread( 'HPRTE2.xlsx' ,1, 'DI11:DI11' ); m_10_gas=1+Recirc; %[kg] ratio=0.3; Conc_EG=0.6; Conc_W=0.4; m_8_EG=m_10_gas*ratio*Conc_EG; %[kg] m_8_w=m_10_gas*ratio*Conc_W; %[kg] T_10=276.2; %[K] P_10=350; %[kpa] P_3=200; %[kpa] deltaT_y=7; %[K] T_def=2; %[K] T_amb=303.15; %[K] eta=0.8; ksi=0.9; phi=1; T_8=250; %[K] %Assumptio ns Q_s=0; W_s=0; v_1=0; v_11=0; v_10=0; v_9=0; z_1=0; z_11=0; z_10=0; z_9=0; m_11_EG=0; Q_hex=0; W_hex=0; W _heater=0; Q_sep=0; W_sep=0; W_amb=0; W_cold=0;

PAGE 66

66 m_3_vapor_EG=0; Q_pump=0; m_5_EG=0; % Another used constants g=9.81; error=1.45; difference=5; %Pressures on the system P_1=P_10; P_2 =P_10; P_9=P_10; P_11=P_10; P_4=P_3; P_5=P_3; P_6=P_3; P_7=P_3; P_8=P_3; % ---------------------------------------------------------%PUMP Function [m_9_EG,m_9_w,T_9,m_8] = mpump(Q_pump,m_8_EG,m_8_w,T_8); %Spray Chamber Function [T_1,T_11,m_1_EG,m_1_w,m_ 10_w,m_1,x_1,C_p_11] = ... mspraychamber(Q_s,W_s,T_9,m_9_EG,m_9_w,T_10,m_10_gas,m_11_EG); %Multi Component Function [T_3,T_6,Q_heater,T_4,T_2,m_6_EG,m_6_w,T_5,x_2,T_5_sat,x_4,m_5_wvap, ... h_3_q,h_3_0,T_4_0,q_3,m_5_EGvap] =mmulticomp(m_1_EG,m_1_w, m_10_w,m_5_EG,T_1,Q_sep,W_sep); %Ambiant Heat Exchanger function [Q_amb,T_7,m_7_EG,m_7_w] =mambheatexc(m_6_EG,m_6_w,W_amb,T_6); %Evaporator Function [Q_cold] =mevap(m_7_EG,m_7_w,T_8,T_7,W_cold); %_ -------------------------------------------------------xlswrite( 'HPRTE2.xlsx' ,Q_heater,1, 'V19:V19' ); xlswrite( 'HPRTE2.xlsx' ,Q_cold,1, 'AK18:AK18' ); xlswrite( 'HPRTE2.xlsx' ,T_8,1, 'AF18:AF18' ); xlswrite( 'HPRTE2.xlsx' ,T_7,1, 'AE18:AE18' ); xlswrite( 'HPRTE2.xlsx' ,T_11,1, 'M20:M20' ); xlsread( 'HPRTE2.xlsx' ,1, 'AL18:A L18' ) %To change variables from MATLAB : xlswrite( 'HPRTE2.xlsx' ,2103.1,1, 'C11:C11' ); Pump function [m_9_EG,m_9_w,T_9,m_8] = mpump(Q_pump,m_8_EG,m_8_w,T_8) %Pump global eta P_8 P_9 m_9_EG=m_8_EG ; m_9_w=m_8_w; m_9=m_9_EG+m_9_w; m_8=m_8_EG+m_8_w;

PAGE 67

67 x_8=m_8_E G/m_8; D_8=densofEGW(x_8); %[kg/m^3] v_8=1/D_8; C_98=spec_heat(x_8,T_8); T_9=T_8+Q_pump/(m_9*C_98)+v_8*(P_9 P_8)/(eta*m_9*C_98); %Correction T_ave_98=(T_9+T_8)/2; C_98=spec_heat(x_8,T_ave_98); T_9=T_8+Q_pump/(m_9*C_98)+v_8*(P_9 P_8)/(eta*m_9*C_98); end Sp ray Chamber function [T_1,T_11,m_1_EG,m_1_w,m_10_w,m_1,x_1,C_p_11] = ... mspraychamber(Q_s,W_s,T_9,m_9_EG,m_9_w,T_10,m_10_gas,m_11_EG) global phi v_1 v_9 v_11 v_10 P_10 g z_1 z_11 z_10 z_9 %Spray Chamber P_sat=REFPROPm( 'P' 'T' ,T_10, 'Q' ,0.5, 'water' ); m_ 10_w=0.622*(phi*P_sat)/(P_10 phi*P_sat)*m_10_gas; m_11_gas=m_10_gas; m_1_EG=m_9_EG m_11_EG; m_1_w=m_9_w+m_10_w; m_9=m_9_EG+m_9_w; m_1=m_1_EG+m_1_w; x_9=m_9_EG/m_9; x_1=m_1_EG/m_1; % -----------------------------------------------C_p_10=1.0035; C_p_11=C_p_ 10; C_w=spec_heat(0,T_10); C_91=spec_heat(x_9,T_9); h_10_vap=REFPROPm( 'H' 'T' ,T_10, 'Q' ,1, 'water' )/1000; h_10_sat=REFPROPm( 'H' 'T' ,T_10, 'Q' ,0, 'water' )/1000; Q_hfg=m_10_w*(h_10_vap h_10_sat); %Assuming T_11=T_1 T_1=(Q_s W_s+Q_hfg+m_9*C_91*T_9+m_10_w*C_w*T_10 +m_11_gas*C_p_11*T_10 ... m_9*((v_1^2)/2 (v_9^2)/2+g*z_1 g*z_9) m_10_w*((v_1^2)/2 ... (v_10^2)/2+g*z_1 g*z_10) m_11_gas*((v_11^2)/2 (v_10^2)/2 ... +g*z_11 g*z_10))/(m_9*C_91+m_10_w*C_w+m_11_gas*C_p_11); T_ave_111=(T_1+T_9)/2; T_ave_101=(T_10+T _1)/2; C_91=spec_heat(x_9,T_ave_111); C_w=spec_heat(0,T_ave_101); T_1=(Q_s W_s+Q_hfg+m_9*C_91*T_9+m_10_w*C_w*T_10+m_11_gas*C_p_11*T_10 ... m_9*((v_1^2)/2 (v_9^2)/2+g*z_1 g*z_9) m_10_w*((v_1^2)/2 ... (v_10^2)/2+g*z_1 g*z_10) m_11_gas*((v_11^2)/2 (v _10^2)/2 ... +g*z_11 g*z_10))/(m_9*C_91+m_10_w*C_w+m_11_gas*C_p_11); T_11=T_1; end

PAGE 68

68 Multi Component function [T_3,T_6,Q_heater,T_4,T_2,m_6_EG,m_6_w,T_5,x_2,T_5_sat,x_4,m_5_wvap, ... h_3_q,h_3_0,T_4_0,q_3,m_5_EGvap] =mmulticomp(m_1_EG,m_1_w,m_10 _w,m_5_EG,T_1,Q_sep,W_sep) %multicomponentsystem global error difference P_4 P_2 P_5 ksi W_heater P_3 m_2_EG=m_1_EG; m_2_w=m_1_w; m_1=m_1_w+m_1_EG; m_2=m_2_EG+m_2_w; x_2=m_2_EG/m_2; %CV 3 m_3_EG=m_2_EG; m_3_w=m_2_w; m_3=m_2; x_3=x_2; %CV 8 m_5_w =m_10_w; m _5=m_5_w+m_5_EG; %CV 4 m_4_EG=m_3_EG m_5_EG; m_4_w=m_3_w m_5_w; m_4=m_4_w+m_4_EG; x_4=m_4_EG/m_4; %CV 2 m_6_EG=m_4_EG; m_6_w=m_4_w; m_6=m_6_EG+m_6_w; x_6=x_4; %Initialize T_4 T_4_0_app=REFPROPm( 'T' 'P' ,P_3, 'Q' ,0, 'eglycol' 'water' ,[x_3 1 x_3]); T_4_defined= T_4_0_app 10; T_4(1)=T_4_defined ; i=1; %Calculating necessary quality value: m_3_vapor_w =m_5_w; q_3_nec=m_3_vapor_w/m_3; q_3=q_3_nec; C_41=spec_heat(x_4,(T_4(i)+T_1)/2); C_21=C_41; %First assuming T_2is close to T_4 while difference > error d elta_h_max=C_21*(T_4(i) T_1); T_2=T_1+ksi*(m_1/(m_1*C_21))*delta_h_max; T_ave_21=(T_1+T_2)/2; C_21=spec_heat(x_2,T_ave_21); delta_h_max=C_21*(T_4(i) T_1); T_2=T_1+ksi*(m_1/(m_1*C_21))*delta_h_max; T_3=REFPROP m( 'T' 'P' ,P_3, 'Q' ,q_3, 'eglycol' 'water' ,[x_3 1 x_3]); T_5=T_3; h_5=REFPROPm( 'H' 'T' ,T_5, 'P' ,P_5, 'water' )/1000; T_5_sat=REFPROPm( 'T' 'P' ,P_5, 'Q' ,0, 'water' );

PAGE 69

69 [liquid vapor]=REFPROPm( 'X' 'P' ,P_3, 'Q' ,q_3, 'eglycol' 'water' ,[x_3 1 x_3]); m_5_wvap=vapor(2)*q_3*m_3; m_5_EGvap=vapor(1)*q_3*m_3; h_5_EG=REFPROPm( 'H' 'T' ,T_5, 'P' ,P_5, 'eglycol' )/1000; h_3_N=REFPROPm( 'H' 'P' ,P_3, 'Q' ,q_3, 'eglycol' 'water' ,[x_3 1 x_3])/1000; h_4_N=Q_sep/m_4 W_sep/m_4+(m _3/m_4)*h_3_N ((m_5_wvap)/m_4)*h_5 (m_5_EGvap/m_4)*h_5_EG; %Concentration changes !! T_4(i+1)=REFPROPm( 'T' 'P' ,P_4, 'H' ,h_4_N*1000, 'eglycol' 'water' ,[x_4 1 x_4]); difference= abs(T_4(i+1) T_4(i)); i=i+1; end T_4=T_4(i); T_3_0=REFPRO Pm( 'T' 'P' ,P_3, 'Q' ,0, 'eglycol' 'water' ,[x_3,1 x_3]); C_32=spec_heat(x_3,(T_3_0+T_2)/2); h_3_q=REFPROPm( 'H' 'P' ,P_3, 'Q' ,q_3, 'eglycol' 'water' ,[x_3,1 x_3])/1000; h_3_0=REFPROPm( 'H' 'P' ,P_3, 'Q' ,0, 'eglycol' 'water' ,[x_3,1 x_3])/1000; Q_heater=m_3*(h_3_q h_3_0) +m_3*C_32*(T_3_0 T_2)+W_heater; C_46=spec_heat(x_4,(T_4+T_1)/2); T_6=T_4 ksi*(m_1/(m_4*C_46))*delta_h_max; T_ave_46=(T_6+T_4)/2; C_46=spec_heat(x_4,T_ave_46); T_6=T_4 ksi*(m_1/(m_4*C_46))*delta_h_max; end Ambient Heat Exchanger function [Q_amb,T_7,m_7_E G,m_7_w] =mambheatexc(m_6_EG,m_6_w,W_amb,T_6) %Ambie nt Heat Exchanger global deltaT_y T_def T_amb P_6 m_7_EG=m_6_EG; m_7_w=m_6_w; m_6=m_6_EG+m_6_w; m_7=m_7_EG+m_7_w; x_6=m_6_EG/m_6; x_7=x_6; T_6_0=REFPROPm( 'T' 'P' ,P_6, 'Q' ,0, 'eglycol' 'water' ,[x_6,1 x_6]); if T_6>T_6_0 T_7=T_amb; %Here we have to think two fictitious heat exchangers again. h_6_val=REFPROPm( 'H' 'P' ,P_6, 'T' ,T_6, 'eglycol' 'water' ,[x_6,1 x_6])/1000; h_6_0=REFPROPm( 'H' 'P' ,P_6, 'Q' ,0, 'eglycol' 'water' ,[x_6,1 x_6])/1000; T_ave_6 7=(T_7+T_6_0)/2; C_76=spec_heat(x_7,T_ave_67); Q_amb=W_amb+m_6*(h_6_0 h_6_val)+m_7*C_76*(T_7 T_6_0); %Genel entalpi aliyoruz NIST den. else %T_6
PAGE 70

70 else T_7=T_6; end T_ave_76=(T_7+T_6)/2; C_76=spec_heat(x_7,T_ave_76); Q_amb=m_7*C_76*(T_7 T_6); end end Evaporator function [Q_cold] =mevap(m_7_EG,m_7_w,T_8,T_7,W_cold) %Evaporator m_7=m_7_EG+m_7_w; x_7=m_7_EG/m_7; T_ave_78=(T_7+T_8)/2; C_78=spec_heat(x_7 ,T_ave_78); Q_cold=W_cold+m_7*C_78*(T_8 T_7); end Specific Heat Calculator function y = spec_heat( x,T ) %global T_4_0 %spec_heat (finding specific heat) %T (temperature)(celcius) %y (kj/(kg.K)) %x EG,weight percent in water %Data taken from MEGlobal da ta sheet for EG Water mixtures. T=T 273; x=x*100; if 0<=x & x<=5 A=1.0038; B= 2.2459e 4; C=2.6257e 6; end if 5
PAGE 71

71 B=5.1554e 4; C=0; end if 35
PAGE 72

72 APPENDIX B BASIC CHEMICAL EQUATIONS FOR COMBUSTION The equivalence ration is given in Turns [10] as a function of fuel and air ratios of stoichometric and real cases to indicate whether a fuel oxidizer mixture is rich, lean or stoichometric. (B 1) By using the equivalence ration definition we can relate this to the amount of excess air we use. (B 2) The mole fraction of any species i in the mixture can be written as the ratio of the mole of species divided by the total number of moles in the system. (B 3) Combustion of the products in recirculated system can be written as following: (B 4) (B 5) (B 6) At the mixing point where fresh air and recirculation flows

PAGE 73

73 (B 7) (B 8) (B 9)

PAGE 74

74 LIST OF REFERENCES [1] Khan, J. R., Lear W. E. Sherif S. A. Crittenden J. F. 2008, "Performance of a Novel Combined Cooling and Power Gas Turbine With Water H arvesting ASME, Journal of Engineering for Gas Turbine and Power, 130, 041702 [2] Choon J. R. 2011, Modeling and Control of A Novel Semi Closed Gas Turbine Absorption Combined Cycle, Dissertation, University Of Florida, Dept. Of Mechanical Engin eering Gainesville Fl [3] Temp: A Novel Web Based Approach for the Thermoeco nomic Analysis and Optimization of Conventional and Innovative Cycles Paper No. 2004 GT 54115 [4] Chen S. Y. Soriano A. N. Li M. H. 2010, "De nsities and Vapor Pressures of M ixed S olvent D esiccant S ystems C ontaining { G lycol ( D iethylene, or T rie thylene, or T etraethylene G lycol)+ S alt ( M agnesium C hloride )+ W ater}," The Journal of C hem. Thermodynamics 42 pp 1163 1167 [5] Trimble H. M. Potts W., 1935, "Gl ycol Water Mixtures Vapor Pressure Boiling Point Composition Relations," Ind. Eng. Chem 27, pp 66 68 [6] Worswick R. D. Dunn A. G. Stavaley L. A. K, 1974, "The Enthalpy of Solution of Ammonia in Water and in Aqueous Solutions of Ammonium Chloride and Ammonium Bromide," The Journal of Che m. Thermodynamics, 6, pp 565 570. [7] Khan J. R. Lear W. E., Sherif S. A. Howell E. B. Crittenden J. F., Meitner P. L., 2010 "A Novel Pressurized CHP System with Water Extraction and Refrigeration Applied Th ermal Engineering, 30, pp. 1081 1090. [8] Boza J. J. 2003, "Performance of A Semi Cl osed Gas Turbine and Absorption Refrigeration Combined Cycle," Master's Thesis University Of Florida, Dept. of Mechanical Engineering, Gainesville Fl [9] Nemec T. S., 1995, "Thermodynamic Design Point Study Of A Semi Closed Recuperated Intercooled Gas Turbine Com bined With A Rankine Bottoming Cycle," Master's Thesis, University Of Florida, Dept of Mechanical Engineering Gainesville, Fl [10] Turns S. R., 2000, An Introduction to Combustion: Concepts and Applications Second Edition McGraw Hill New Y ork, USA. [11] Cengel Y. A., 2003, Heat Transfer: A Practical Approach 2nd Edition McGraw Hill New York, USA. [12] Lemmon E. W., 2009, Equation of S ta te Fitting with REFPROP P rogram Physical and Chemical Properties Division, NIST, Colorado, USA

PAGE 75

75 [13] ME Global Company Ethylene Glycol Product Guide USA Available: http://www.meglobal.biz/media/product_guides/MEGlobal_MEG.pdf [14] Kracht C., Ulbig P., Schulz S., 1999, "Measurement and Correlation of Excess Molar Enthalpies for (Ethendiol, or 1,2 P ro panediol, or 1,2 Butanediol + Water) at T emperatures (285.65, 298.15, 308.15, 323.15, and 338.15)K," The Journal of C hem. Thermodynamics, 31, pp. 1113 1127 [15] Perry H. R., 1997, Ch emical Engineer's Handbook 7th E dition McGraw Hill New York, USA. [16] Boyce, Mehervan P., 2006, Gas Turbine Engineering Handbook (3rd Edition) Elsevier, ISBN 978 0 7506 7846 9 [17] Anxionnaz, R., 1945, Installation a Turbines a Gaz a Circuit Semi O uvert French Patent 999 133. [18] Anxionnaz, R., 1948 Improvements in or Relating to Gas Turbine Plant with Semi Open C ircuit British Patent 651 166. [19] Lear, W. E., Laganelli, A. L., 1999, High Pressure Regenerative Turbine E ngine: 21 st Century P Report for Contract No. NAS3 27396. [20] Gasparovi c, N., 1965, The A dvantage of Semi Closed Cycle Gas Turbines for Naval S hip P ropulsion 77, pp. 275 333. [21] Fiaschi D., Manfrida G., 1999, A New Semi Closed Gas Turbine C ycle with CO 2 S eparation Energy C onversion and Management, 40, pp. 1669 1678. [22 ] Howell, E. B., 2007, "Experimental Study of a Novel G as Turbine Engine I ntegrated with an Absorption Refrigeration S ystem, Master's Thesis, University Of Florida, Dept. o f Mechanical Engineering, Gainesville, Fl [23 ] Tiffany, D. R., Lear, W., E., Harris, M., A, Crittenden, J., F., 2009, "Design of a Novel Quad G ener ation D istribu ted Energy Demonstration P Paper No. 2009 GT 60312, pp. 311 318. [24 ] http://www.mhi.co.jp/en/news/story/1105261435.html

PAGE 76

76 BIOGRAPHICAL SKETCH Erman K. Oztekin was born in Edirne, Turkey in May 1986. He graduated from Istanbul University Mechanical Engineering Department in 2008. He had an internship in ANNOVA engineering for two months. He joined Dr. William E. Lear's group in 2010 as a master' s student.