- Email: [email protected]
Experimental results and thermodynamic analysis of a natural gas small scale cogeneration plant for power and refrigeration purposes
Accepted Manuscript Experimental results and thermodynamic analysis of a natural gas small scale cogeneration plant for power and refrigeration purposes Edson Bazzo, Alvaro Nacif de Carvalho, José Alexandre Matelli PII:
S1359-4311(13)00312-8
DOI:
10.1016/j.applthermaleng.2013.04.041
Reference:
ATE 4775
To appear in:
Applied Thermal Engineering
Received Date: 19 November 2012 Accepted Date: 25 April 2013
Please cite this article as: E. Bazzo, A. Nacif de Carvalho, J.A. Matelli , Experimental results and thermodynamic analysis of a natural gas small scale cogeneration plant for power and refrigeration purposes, Applied Thermal Engineering (2013), doi: 10.1016/j.applthermaleng.2013.04.041. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights of the manuscript entitled
EXPERIMENTAL RESULTS AND THERMODYNAMIC ANALYSIS OF A NATURAL GAS SMALL SCALE COGENERATION PLANT FOR POWER AND REFRIGERATION PURPOSES
RI PT
by Edson Bazzo, Alvaro N. Carvalho and José A. Matelli
A small scale cogeneration plant for power and refrigeration is proposed and analyzed.
•
The plant is based on a microturbine and a modified absorption chiller;
•
The plant is analysed based on 1st and 2nd laws of thermodynamics.
•
Experimental results are found for different power and refrigeration conditions.
•
The plant proved to be technically feasible.
AC C
EP
TE D
M AN U
SC
•
ACCEPTED MANUSCRIPT
RI PT
A manuscript entitled:
M AN U
by:
SC
EXPERIMENTAL RESULTS AND THERMODYNAMIC ANALYSIS OF A NATURAL GAS SMALL SCALE COGENERATION PLANT FOR POWER AND REFRIGERATION PURPOSES
Edson Bazzo1
Alvaro Nacif de Carvalho1 José Alexandre Matelli2†
TE D
† Author for correspondence [email protected] Phone: +55 12 3123-2839 fax: +55 12 3123-2835
AC C
EP
1. UFSC – Federal University of Santa Catarina Department of Mechanical Engineering 88040-900 Florianópolis, SC – Brazil 2. UNESP – Univ Estadual Paulista Department of Energy 12516-410 Guaratinguetá, SP – Brazil
To be submitted to the Applied Thermal Engineering
Guaratinguetá, winter 2012.
1
ACCEPTED MANUSCRIPT
Abstract. In this work, experimental results are reported for a small scale cogeneration plant for power and refrigeration purposes. The plant includes a natural gas microturbine and an ammoniawater absorption chiller fired by steam. The system was tested under different turbine loads, steam
RI PT
pressures and chiller outlet temperatures. An evaluation based on the 1st and 2nd Laws of Thermodynamics was also performed. For the ambient temperature around 24°C and microturbine at full load, the plant is able to provide 19 kW of saturated steam at 5.3 bar (161ºC), corresponding to 9.2 kW of refrigeration at –5°C (COP = 0.44). From a 2nd law point-of-view, it was found that
SC
there is an optimal chiller outlet temperature that maximizes the chiller exergetic efficiency. As expected, the microturbine presented the highest irreversibilities, followed by the absorption chiller
M AN U
and the HRSG. In order to reduce the plant exergy destruction, it is recommended a new design for the HRSG and a new insulation for the exhaust pipe.
AC C
EP
TE D
Keywords. Cogeneration, Microturbine, Absorption chiller, Natural gas.
2
ACCEPTED MANUSCRIPT
Nomenclature c
specific exergy cost [USD/kWh]
Subscripts 0
reference state
coefficient of peformance [–]
b
brine
C˙
exergy cost rate [USD/h]
d
destruction
e
specific exergy [kJ/kg]
e
electric
E˙
exergy rate [kW]
f
fuel
f
exergoeconomic factor [–]
g
exhaust gases
h
specific enthalpy [kJ/kg]
i
inlet stream
LHV
lower heat value [kJ/kg]
j
j-th stream
m ˙
mass flow [kg/s]
k
k-th plant component
P
pressure [kPa]
l
lost
˙ Q
heat transfer rate [kW]
r
relative cost difference [–]
s
specific entropy [kJ/kgK]
T
temperature [°C]
W˙
work transfer rate [kW] exergy destruction ratio [–]
Z˙
purchase cost rate [USD/h]
Greek letters
TE D
y
energy-based efficiency [–]
ε
exergy-based efficiency [–]
natural gas
o
outlet stream
p
product
q
heat related
s
steam
w
work related
Acronyms AC
absorption chiller
EP
exhaust pipe
HRSG heat recovery steam generator MT
microturbine
AC C
EP
η
SC
ng
M AN U
COP
RI PT
specific heat [kJ/kgK]
cP
3
ACCEPTED MANUSCRIPT
1. Introduction
The use of waste heat as energy source for absorption chillers can be an alternative to decrease
RI PT
operating costs of either air conditioning or refrigeration processes. Usually, the waste heat comes from gas turbines, reciprocating engines or from some other industrial processes. The use of
cogeneration systems based on combined cooling, heating and power plants (CCHP) in the industry [1, 2] and in commercial buildings [3, 4] was investigated and it is well documented in the
SC
literature.
Recent results related to experimental and theoretical investigations of small scale cogeneration
M AN U
plants based on microturbines and absorption chillers are reported [5–10], demonstrating the technical feasibility and the potential to achieve high overall performance. Aiming to assure the feasibility of these systems, several heat recovery technologies for refrigeration are recommended for different levels of temperature [11]. The influence of the heat source temperature on COP and
TE D
cooling output of H2O-LiBr and H2O-NH3 absorption chillers are also investigated [12–16]. Recently, Rossa and Bazzo [17] reported a thermodynamic first and second law analysis of a small scale cogeneration plant for power and air-conditioning. The plant consists of a 28 kWe
EP
natural gas microturbine, a thermosyphon heat exchanger (the heat from the microturbine exhaust gases is recovered to generate hot water) and a H2O-NH3 absorption chiller using hot water as heat
AC C
source.
The objective of the present paper is to report the experimental data of an upgraded version of the plant studied by Rossa and Bazzo [17]. The new plant is conceived for refrigeration purposes and consists of a 28 kWe natural gas microturbine, a HRSG (the heat from the microturbine exhaust gases is recovered to generate saturated steam) and a H2O-NH3 absorption chiller using saturated steam as heat source. Instruments for pressure, mass flow rates and temperature were placed in key points of the plant. The tests were carried out for different levels of power output power, steam pressure and chiller output temperature, with the respective experimental results reported in this paper. In order to identify possible improvements in the plant performance, a thermodynamic 4
ACCEPTED MANUSCRIPT
analysis is held. Some changes are proposed to reduce the exergy destruction and to increase the overall efficiency.
RI PT
2. Small scale cogeneration plant description
The schematic diagram of the small scale cogeneration plant is shown in Fig. 1. A set of six
cylinders store and supply natural gas to the microturbine. The heat from the turbine exhaust gases
SC
is recovered in the HRSG, where steam is generated in order to feed the H2O-NH3 chiller. The cooling output from the chiller is a brine solution of monoethyleneglycol, which is stored in a
M AN U
proper tank. The tank has electrical resistors to emulate a thermal load. In addition, volumetric flow rate devices as well as temperature and pressure sensors are all connected to a data acquisition system.
TE D
Figure 1. Scheme of the small scale cogeneration plant.
The prime mover is a Capstone C30 LP microturbine. It includes a compressor, a regenerator, a
EP
combustion chamber, a turbine and a generator. The rotating components are all assembled over a single shaft supported by air bearings and can reach angular velocities as high as 96000 rpm. There
AC C
is no reduction gearbox between the generator shaft and the gas turbine shaft, so that the generator power output is electronically treated in order to provide proper values of AC voltage and frequency. Operating at ISO condition (15°C and 101.3 kPa), the microturbine generates 28 kW of power and 84 kW of recoverable heat from 0.31 kg/s of exhaust gases at 275 °C, with an electrical efficiency equal to 0.25 [18]. The HRSG consists of a diverter valve, a cross-flow heat exchanger and a steam drum. The residual heat from the microturbine is recovered into the HRSG to generate saturated steam for the chiller. For security reasons, the HRSG is designed for a maximum pressure of 6 bar. The heat input to the HRSG is automatically controlled by an on-off diverter valve, which acts according the steam 5
ACCEPTED MANUSCRIPT
pressure signal. Thus, the exhaust gases are either diverted directly to the atmosphere to avoid over pressure or flows through the heat exchanger to keep the pressure between specified values. The absorption chiller is a Robur ACF60-00 LB designed for refrigeration, with cooling
RI PT
capacity of 13.3 kW and output temperature of –5°C. Additional performance data of the chiller are presented in Tab. 1. These figures are rated by the manufacturer at an ambient temperature of 35°C [19]. The manufacturer provides a correction factor (taken into account in this work) to predict the chiller performance at different ambient temperatures and different cooling temperatures, as shown
SC
in Tab. 2 [19].
M AN U
Table 1. Chiller performance data (adapted from [19]).
Table 2. Correction factor for the cooling capacity (adapted from [19]).
The chiller is a single-block unit based on a H2O-NH3 solution and single effect cycle with an
TE D
air-cooled condenser. It is designed to work with direct natural gas firing through a burner. The heat from the combustion is transferred to the chiller generator. Since the heat is transferred mainly by convection, the generator has fins in its external surface (Fig. 2a). In order to make the chiller
EP
suitable for cogeneration, the burner was removed and a properly designed jacket was installed. The generator is now fully enveloped by the jacket, so that the steam enters the jacket, flows around the
AC C
external generator fins and leaves the jacket as condensate (Fig. 2b). When a steady-state condition is established during a particular test, the steam mass flow remains constant. However, the steam mass flow may vary from one steady-state test to another. Since no instrument is installed to measure the steam flow rate entering the absorption chiller, an energy balance in the HRSG is considered to estimate the steam mass flow that condenses in the generator jacket. Other important modification was made in the chiller control system, originally designed to control the heat input in the chiller generator according the cooling outlet temperature. Basically, this system was shutdown and replaced by a customized one. Since the outlet cooling temperature and the steam pressure are
6
ACCEPTED MANUSCRIPT
related, the steam pressure signal is used to control the HRSG diverter valve and, consequently, the heat input to the chiller generator. The chiller cooling output is a brine solution of water-
RI PT
monoethyleneglycol (33% in mass) able to reach temperatures below –10°C without freezing.
Figure 2. Chiller generator: (a) original design; (b) modified for cogeneration.
SC
The brine circulates continually between the tank and the chiller. Four electrical resistors of 4.8 kW each are immersed in the tank in order to emulate a thermal load and to allow steady state plant
M AN U
operation.
A particularity of this plant is that the microturbine is placed indoors and all other components are placed outdoors. Thus, the microturbine and the chiller can be subject to slightly different
3. Methodology
TE D
ambient temperatures.
EP
3.1. Plant experimental tests
For the power output, the measurements of power, voltage and current can be taken directly
AC C
from the microturbine control system. For the cooling output, measurements of temperatures, pressures and flow rates in strategic points of the plant are used to obtain thermodynamic properties for further evaluation. These measurements are available from the data acquisition system. In order to evaluate the influence of the steam pressure on the chiller performance, the microturbine was set to different power outputs. For each power output there is a corresponding exhaust gases flow and temperature. Hence, for each power output there is a corresponding steam pressure in the HRSG. Since the tests were carried out during several days, the weather forecast was checked in order to assure some repeatability in the ambient temperature.
7
ACCEPTED MANUSCRIPT
The tests were performed at steady state condition for different power outputs: 16, 18, 20, 22, 24 and 26 kW. When the pressure was stable in the HRSG, the absorption chiller worked with different cooling outputs by setting the brine temperature to 0, –2, –5, –7 and –9.5 °C (point 8, see Fig. 1).
RI PT
During the tests, the ambient temperature ranged between 22.1°C and 24.5°C. Although the nominal cooling output and COP were rated by the chiller manufacturer at 35°C, an available correction factor (Tab. 2) was used for different ambient temperatures.
SC
3.2. Thermodynamic analysis
M AN U
A thermodynamic analysis is performed in order to evaluate the plant performance and consolidate the results from the experiments. The analysis considers four subsystems: the microturbine (MT), the heat recovery steam generator (HRSG), the absorption chiller (AC) and the exhaust gases pipe from the microturbine to the steam generator (EP). The EP subsystem includes
TE D
the diverter valve. Since the exhaust gases pipe is relatively long, the EP subsystem is considered to evaluate the exhaust gas losses between the microturbine outlet and the steam generator inlet. This analysis focuses on the identification of the irreversibilities. The cost contribution of each
EP
subsystem and an exergoeconomic analysis are found in the Appendix A. The following assumptions are considered in the analysis:
• • •
Steady state condition;
AC C
•
Exhaust gases properties as air properties; Ideal gas model;
Negligible kinetic and potential energy changes;
•
Saturated steam (x=1) at HRSG outlet;
•
Saturated liquid (x=0) at HRSG inlet;
•
Negligible liquid pumping power consumption;
•
Negligible pressure drop through the HRSG;
8
ACCEPTED MANUSCRIPT •
For simplicity, the HRSG efficiency is constant and equal to 0.92.
Mass, energy and exergy balances are considered for each subsystem, taking into account heat
RI PT
transfer, work interaction and the inlet/outlet material streams, as shown in Eqs. 1–3. The detailed set of equations is shown in Appendix B. The thermodynamic properties are obtained from the Engineering Equation Solver library. All the equations refers to the points depicted in Fig. 1.
SC
∑ m˙ i =∑ m˙ o
E˙ q −W˙ =∑ m ˙ o e o −∑ m˙ i ei + E˙ D
M AN U
˙ W˙ =∑ m˙ o ho −∑ m˙ i hi Q−
(1) (2) (3)
Considering the absence of entropy data for the brine solution of monoethyleneglycol, the corresponding exergy was estimated by Eq. B16. The First Law-parameters described in Eqs. 4–7
η e=W˙ e / ( m ˙ 2 LHV )
Q˙ s= m ˙ 6 ( h 6−h7 )
EP
η ov=( W˙ e+ Q˙ s ) / ( m˙ 2 LHV )
TE D
are considered to evaluate the performance of the plant:
AC C
COP= [ m ˙ 8 c P,b ( T 9 −T 8 ) ] / [ m ˙ 4 c P,g ( T 4 −T 5 ) ]
(4) (5) (6) (7)
Based on the natural gas composition informed by the local gas utility, the lower heating value and the exergy of the fuel are calculated as 47308 kJ/kg and 49437 kJ/kg, respectively. The exergy of the fuel components is given in Kotas [20]. The overall efficiency in Eq. 5 is stated considering the steam input in the chiller instead of its cooling output. In this way, it is possible to evaluate the maximum amount of fuel energy that is effectively converted into power or steam for comparison with other cogeneration systems. So, the absorption chiller is considered as a subsystem that uses
9
ACCEPTED MANUSCRIPT
the available steam as energy source. Thus, the COP (Eq. 7) is the parameter that completes the plant evaluation. The parameters shown in Eqs 8–12 are considered for evaluation based on the Second Law, on
RI PT
which the irreversibilities and exergy destructions are identified. For the microtubine in particular, both exergy destruction and exery lost were considered in the definition of its exergy efficiency. The exergy destruction ratio of a subsystem k, as shown in the Eq. 12, represents the contribution of
SC
exergy destruction of a subsystem related to the total exergy invested in the system.
ε HRSG =[ m ˙ 6 ( e 6− e 7 ) ] / [ m ˙ 4 ( e 4 −e 5 ) ] ε AC=[ m ˙ 8 ( e 9−e8 ) ] / [ m ˙ 6 ( e 6−e 7 ) ]
˙ e +m ε ov =[ W ˙ 6 ( e 6 −e 7 ) ] / ( m ˙ 2 e ng )
(8) (9) (10) (11) (12)
EP
4. Results
TE D
y d,k = E˙ k / ( m ˙ 2 e ng )
M AN U
˙ e / ( m˙ 2 e ng ) ε MT =W
AC C
4.1. Experimental tests results
The steam pressure obtained in the HRSG for different microturbine power output is shown in Fig. 3. As expected, the higher the power output, the higher the steam pressure obtained in the HRSG. It is important to remind that a steady state condition was assured for each power output. The linear behavior presented in Fig. 3 is also expected, since the microturbine exhaust gases heat increases linearly with power output. In addition, the steam generation is a thermodynamic process at constant volume. In such a process, the heat transferred from the turbine exhaust gases increases
10
ACCEPTED MANUSCRIPT
the steam enthalpy. Since the pressure variation is approximately linear with the steam enthalpy, the steam pressure increases linearly with turbine power output.
RI PT
Figure 3. Variation of the steam pressure with power.
For different brine outlet temperatures, the cooling capacity variation with steam pressure is
shown in Figure 4. Again, a linear trend can be observed for the different brine temperatures. It is
SC
explained as follows: the ammonia-water evaporation in the chiller generator is a process at constant volume. In addition, the mass balance in the generator requires that the ammonia mass
M AN U
fraction remains constant. Consider that the heat transferred to the generator increases and both the volume and the ammonia mass fraction remain constant. In this case, the amount of ammonia-water vaporized increases linearly with the heat transferred, resulting in a higher refrigerant mass flow in the evaporator and, consequently, a higher cooling capacity. Since the heat transferred to the
TE D
generator is directly related to the steam pressure, it follows that the cooling capacity increases linearly with the steam pressure.
EP
Figure 4. Variation of the cooling capacity with steam pressure.
AC C
For different steam pressures, the cooling capacity as a function of the brine outlet temperature is shown in Figure 5. As well known in refrigeration systems, the cooling capacity increases linearly with the evaporation temperature, which is also linearly related to the brine temperature in this case. The direct firing points shown in Fig. 5 are calculated values from the manufacturer documentation [19] rather than measured data. These points are presented in order to provide a comparison between the cooling capacity of the modified generator and its original design. At full load (26 kWe), nominal chiller outlet brine temperature (–5°C) and ambient temperature around 24 °C, the absorption chiller provided 9.2 kW of cooling capacity. Considering the correction factor for
11
ACCEPTED MANUSCRIPT
25 °C presented in Tab. 2, the direct firing system provides 14.9 kW of cooling capacity for the same brine temperature. Thus, a reduction of 38% in the cooling capacity is observed.
RI PT
Figure 5. Variation of the cooling capacity with chiller outlet temperature.
Figure 6 shows the variation of the coefficient of performance with steam pressure for different cooling temperatures. The observed trend is that the COP remains approximately constant for the
SC
pressure range tested. This result is in accordance with the results from the work of Fernandéz-
Seara and Vásquez [15], on which the authors reported no significant changes in the COP when the
M AN U
heat source temperature of a single effect ammonia/water absorption chiller ranged from 138°C (2.4 bar) to 161°C (5.3 bar).
Figure 6. Variation of the coefficient of performance with steam pressure.
Figure 7 shows the COP variation with chiller outlet temperature. Again, it is well known that
TE D
the COP increases linearly with evaporation temperature. When fed by steam, the chiller presents a COP reduction relative to its original design. At full load (26 kWe), nominal refrigeration temperature (–5°C) and ambient temperature around 24 °C, a COP of 0.44 was found. That means
EP
25% less if compared to the nominal value of 0.59 for the direct firing system. It can be also seen that the performance of the modified chiller is lower than the direct-firing for all tests. It can be
AC C
explained because the chiller generator is originally designed for both convection and radiation heat transfer from a natural gas flame, as long as operating with steam, the heat transfer is mainly from changing phase convection. Additionally, the generator fins impose a high heat transfer resistance when the convection heat transfer coefficient is as high as in changing phase convection. It is important to remind that the COP as presented in Figs 6 and 7 is calculated based on the heat recovered from the exhaust gases.
Figure 7. Variation of the coefficient of performance with chiller outlet temperature.
12
ACCEPTED MANUSCRIPT
Experimental results considering the 2nd law are also presented. Figure 8 shows that the chiller exergetic efficiency decreases with the steam pressure. This is expected, because the temperature
RI PT
difference in any heat transfer process causes entropy generation, unless the temperature difference tends to zero. A higher steam pressure means a higher temperature difference between the steam and the ammonia-water solution in the generator, so that the entropy generation is higher at high
SC
steam pressures. Thus, the exergy efficiency of the chiller decreases with steam pressure.
M AN U
Figure 8. Variation of the chiller exergetic efficiency with steam pressure.
Figure 9 presents the variation of the chiller exergetic efficiency with cooling temperature for different steam pressures. The results show that, for any steam pressure, the chiller exergetic
TE D
efficiency reaches a maximum value for the refrigeration temperatures around –7°C.
EP
Figure 9. Variation of the chiller exergetic efficiency with cooling temperature.
AC C
4.2. Thermodynamic analysis results
The thermodynamic analysis is held considering the experimental data when the microturbine operated at full load (26 kWe) and ambient temperature equal to 24.4 °C and the chiller at nominal cooling temperature (–5 °C) and ambient temperature equal to 24.8 °C. This slightly difference in the ambient temperature is observed because the microturbine is placed indoors and all other components are placed outdoors. The thermodynamic properties (temperature, pressure and mass flow rate and exergy) related to the material streams are all shown in Tab. 3. The material streams are labeled as shown in Fig. 1.
13
ACCEPTED MANUSCRIPT
Table 3. Thermodynamic properties of each material stream.
RI PT
The energy and exergy related to the energy streams are shown in Tab. 4. The first and second law parameters of performance for each plant subsystem are shown in Tab. 5.
Table 4. Energy and exergy of each energy stream.
SC
Table 5. Performance parameters of each plant subsystem.
M AN U
By far the microturbine presents the higher exergy destruction of the plant, which is expected due to the internal combustion process. Except for pre-heating the reactants or adjusting the air-fuel ratio, nothing can be done to minimize the exergy destruction of a combustion process. Since the microturbine presents an air preheater and its air-fuel ratio is preset, no further actions can be taken
considered.
TE D
in this case. Thus, the replacement of the current microturbine for a more efficient one should be
Regarding the HRSG, the sources of exergy destruction are friction and, most significantly,
EP
stream-to-steam heat transfer. Thus, the main action that could be taken is to optimize the HRSG heat exchange surface in order to obtain pinch point values (∆Tpp = T5 – T6) within practical
AC C
recommended values (10–25°C). A design that minimizes the gas stream pressure drop contributes to reduce the exergy destruction related to the friction. It can be also observed that the HRSG outlet temperature (point 5) indicates that there is still room for heat recovering, since it is well known that this temperature can be as low as 110 °C. The poor insulation in the exhaust pipe between the microturbine and HRSG causes an exergy destruction even higher than the exergy destroyed in the HRSG itself. Thus, a new insulation for the exhaust pipe is recommended. The insulation thickness of a non-isothermal pipe that minimizes the exergy destruction can be determined according the methodology proposed by Bejan et al. [21]. Due to the significant temperature difference between the steam and the ammonia-water solution in 14
ACCEPTED MANUSCRIPT
the generator, there is also exergy destruction in the chiller. The recommendation in this case is a new generator design that maximizes the convection heat transfer from the condensing steam. The results from the thermodynamic analysis are complemented by the results from the
RI PT
exergoeconomic analysis presented in the Appendix A.
5. Conclusion
SC
An experimental study related to a small scale cogeneration plant, consisting of a microturbine, a heat recovery steam generator and an ammonia-water absorption chiller was presented. The chiller
M AN U
was modified in order to use saturated steam as heat source instead the original natural gas directfiring system. The steam is generated from the microturbine exhaust gases heat. The plant proved to be technically feasible for power generation and refrigeration for freezing purposes. For different microturbine power output, the chiller performance was evaluated for different cooling
TE D
temperatures. It is important to emphasize that the tests were carried out during several days. Although weather forecast was checked in order to assure repeatability, some variation in the ambient temperature was inevitability observed. At full load and ambient temperature around 24°C,
EP
the cogeneration plant provided 26 kW of power and 19 kW of saturated steam at 5.3 bar (161ºC), so that 9.2 kW of cooling capacity was generated at –5 °C. The COP was found as 0.44 and the
AC C
overall efficiency was found as 39.4%. Thus, the chiller performance operating in cogeneration is 25% less than the nominal value of 0.59 for the direct firing system. The chiller modification explains this result, because the chiller generator is originally designed for both convection and radiation heat transfer from a natural gas flame, as long as operating with steam, the heat transfer is mainly from changing phase convection. Additionally, the generator fins impose a high heat transfer resistance when the convection heat transfer coefficient is as high as in changing phase convection. However, the HRSG outlet temperature indicates that there is room for heat recovery improvement. Higher steam pressures lead to a substantial increase in the chiller output, although no significant changes were observed in the COP. Both cooling output and COP decreased when 15
ACCEPTED MANUSCRIPT
using the residual heat system with steam, comparing to the direct firing system values available in the manufacturer’s catalogue. From a 2nd law point-of-view, it was found that there is an optimal chiller outlet temperature that maximizes the chiller exergetic efficiency. As expected, the
RI PT
microturbine presented the highest irreversibilities, followed by the absorption chiller and the HRSG. In order to reduce the plant exergy destruction, it is recommended a new design for the HRSG and a new insulation for the exhaust pipe.
SC
Acknowledgements
M AN U
The authors appreciate the FINEP (Research and Projects Financing), the CNPq (National Council for Scientific and Technological Development) and the companies Petrobras, TBG and SCGás, for financial support granted to this research. The authors are also thankful to the Labsolda/UFSC for building the customized control system for the chiller. Part of this research was
AC C
EP
TE D
developed under the cooperation agreement between UFSC and UNESP.
16
ACCEPTED MANUSCRIPT
References
[1] L. Trygg, S. Amiri, European perspective on absorption cooling in a combined heat and power
1337.
RI PT
system – A case study of energy utility and industries in Sweden, Appl. Energy, 84 (2007) 1319–
[2] I. Svensson, B. Moshfegh, System analysis in a European perspective of new industrial cooling supply in a CHP system, Appl. Energy, 88 (2011) 5164–5172.
SC
[3] G.G. Maidment, X. Zhao, S.B. Riffat, G. Prosser, Application of combined heat-and-power and absorption cooling in a supermarket, Appl. Energy, 63 (1999) 169–190
M AN U
[4] G.G. Maidment, R.M. Tozer, Combined cooling heat and power in supermarkets, Appl. Therm. Eng. 22 (2002) 653–665.
[5] S. Tassou, I. Chaer, N. Sugiartha, Y. Ge, D. Marriott, Application of tri-generation systems to the food retail industry, Energy Convers. Manag. 48 (2007) 2988–2995.
TE D
[6] L. Colombo, F. Armanasco, O. Perego, Experimentation on a cogenerative system based on a microturbine, Appl. Therm. Eng. 27 (2007) 705–711. [7] A. Zaltash, A., Petrov, D. Rizy, S. Labinov, E. Vineyard, R. Linkous, Laboratory R&D on
EP
integrated energy systems (IES), Appl. Therm. Eng. 26 (2006) 28–35. [8] M. Medrano, J. Mauzey, V. McDonell, S. Samuelsen, D. Boer, Theoretical Analysis of a Novel
AC C
Integrated Energy System Formed by a Microturbine and an Exhaust Fired Single-Double Effect Absorption Chiller, Int. J. Thermodyn. 9 (2006) 29–36. [9] J. Ho, K. Chua, S. Chou, Performance study of a microturbine system for cogeneration application, Renew. Energy, 29 (2004) 1121–1133. [10] M.S. Rocha, R. Andreos, J.R. Simões-Moreira, Performance tests of two small trigeneration pilot plants, Appl. Therm. Eng. 41 (2012) 84–91. [11] D. Wu, R. Wang, Combined cooling, heating and power: A review, Energy Combust. Sci. 32 (2007) 459–495.
17
ACCEPTED MANUSCRIPT
[12] G. Grossman, Solar-powered systems for cooling, dehumidification and air-conditioning, Solar Energy, 72 (2002) 53–62. [13] M. Izquierdo, E. Palacios, J. Marcos, D. Mendoza, Air cooling by a solar absorption system:
RI PT
2008 summer experimental results”, in: Proceedings of ECOS 2009, Aug. 31 - Sept. 03, 2009, Foz do Iguaçu, Brazil.
[14] A. Keçeciler, H. Acar, A. Dogan, Thermodynamic analysis of the absorption refrigeration
system with geothermal energy: an experimental study, Energy Convers. Manag. 41 (2000) 37–48.
SC
[15] J. Fernández-Seara, M. Vázquez,Study and control of the optimal generation temperature in NH3-H2O absorption refrigeration systems, Appl. Therm. Eng. 21 (2001) 343–357.
Int. J. Refrig. 27 (2004) 10–16.
M AN U
[16] I. Horuz, T. Callander, Experimental investigation of a vapor absorption refrigeration system,
[17] J. A. Rossa, E. Bazzo, Thermodynamic modeling for an ammonia-water absorption system associated to a microturbine, Int. J. Thermodyn. 12 (2009) 38–43.
March 27, 2013.
TE D
[18] Capstone C30 LP. Product catalogue. Available at http://www.capstoneturbine.com, access
[19] Robur ACF 60-00 LB. Installation, use and maintenance instructions (in Italian). Available at
EP
http://www.robur.it, access March 27, 2013.
[20] T. Kotas, The exergy method of thermal plant analysis, Krieger Publishing Company, Malabar,
AC C
1995.
[21] A. Bejan, G. Tsatsaronis, M. Moran, Thermal Design and Optimization, Jonh Wiley & Sons, New York, 1996.
18
ACCEPTED MANUSCRIPT
Appendix A: Exergoeconomic analysis
An exergoeconomic analysis is also performed for understanding the cost formation process and
RI PT
the costs flow in the system, so that potential improvements are identified in order to minimize the final cost of the products. The exergoeconomic model consists of a balance of these costs. Actual purchase and installation costs of the plant in the laboratory were taken into account; fuel and power prices are given by the respective local utilities. Additional expenses were considered as 50% of the
SC
total purchase equipment costs. All costs were converted to US dollars with an exchange ratio of 1.71 R$/US$ at the time this paper was written. The purchase costs of the equipments were
M AN U
levelized based on 20 years of operation, 7446 hours of operation per year, 12% of interest loan and 5% for operation and maintenance costs per year. The inflation rate was not considered. The equations presented in this section were stated in accordance with methodology presented in Bejan et al. [21]. The cost rate for a stream j and the cost balance for a component k are given as the
TE D
following:
C˙ j =c j E˙
j
C˙ w,k + ∑ C˙ o,k = C˙ q,k + ∑ C˙ i,k + Z˙ k o
i
EP
The cost related to the lost and destroyed exergy in a component k depends on the stream that
AC C
carry the lost and destroyed exergy out of the component: C˙ d,k =c l,k ( E˙ d + E˙ l )k
If power and brine are considered as plant final products, the cost flow and the specific cost of the brine are respectively given as follows: ˙ ) + C˙ /2 C˙ b =( C˙ 8 −C 9 5 ˙ c b =C˙ b / W b
The relative cost difference (Eq. A1) measures the relative increase in the specific cost between fuel and product of each subsystem or component. The exergoeconomic factor (Eq. A2) expresses
19
ACCEPTED MANUSCRIPT
the weight of the sources on the cost aggregation, whether it is greater for the subsystem purchase price or for exergy destruction and loss. c p,k − c f,k c f,k
f k=
(A1)
Z˙ k Z˙ k + (C˙ d + C˙ l )
RI PT
r k=
(A2)
SC
The exergoeconomic model is evaluated considering the experimental data when the
microturbine operated at full load (26 kWe) and the chiller at nominal cooling temperature (–5 °C).
M AN U
The cost flow rate and specific exergy cost of the fuel and products are presented in Table A1. The steam cost is applicable only in case of the cogeneration plant generating power and steam as final products. According to the local power utility, the power price is 0.571 US$/kWh for the peak period and 0.084 US$/kWh for the off-peak period. These figures show that the power generated by
TE D
the cogeneration plant is far more expensive than the current market off-peak power prices. On the other hand, it is competitive when compared to the peak power price. Table A2 shows the exergoeconomic parameters regarding the plant subsystems.
AC C
EP
Table A1. Cost flow rate and specific exergy cost of each energy stream (fuel and products).
Table A2. Exergoeconomic parameters of each plant subsystem.
As expected, the microturbine is the subsystem that most contribute for cost aggregation in the system. The microturbine exergoeconomic factor (f) found here is in accordance with the figures found in the literature, as those found in Bejan et al. [20]. The cost is quite well distributed between investment cost and exergy destruction cost. However, the high relative cost difference (r) indicates that cost contribution of the microturbine in the overall system is high.
20
ACCEPTED MANUSCRIPT
The absorption chiller presented the highest relative cost difference due to the low exergy related to its products (streams 8 and 9). Equipments for refrigeration purposes will always aggregate a high cost per exergy unit to the streams. In order to reduce this factor, the chiller should
RI PT
operate at high exergy efficiencies. In the present work, the experimental results show that this operation point corresponds to an outlet brine temperature of –7°C (Fig. 9). The exergoeconomic factor shows that purchase costs and exergy destruction costs are quite balanced. However, the lack of information for cooling equipments in the literature does not allow a more precise analysis.
SC
As the microturbine, the HRSG presented a typical value of exergoeconomic factor. The relative cost factor was found to be high. In this case, it is possible to enhance the exergetic efficiency by
M AN U
using a HRSG with lower pinch point. The exergy destruction in the exhaust pipe causes an increase of 20% on the specific exergy cost. The poor insulation between the microturbine outlet and the HRSG inlet causes an additional cost of 0.38US$/h to the fuel of this subsystem to cover the exergy destruction. A review of the insulation is recommended. In general, it was observed that
efficiency/capacity ratios.
TE D
small scale equipments present a high relative cost factor, due to the high cost/capacity and low
EP
Appendix B: Detailed set of equations of the thermodynamic analysis
AC C
Mass, energy and exergy balances are considered for each subsystem, taking into account heat transfer, work interaction and the inlet/outlet material streams. The subsistems are the microturbine (MT), the heat recovery steam generator (HRSG), the absorption chiller (AC) and the exhaust gases pipe from the microturbine to the steam generator (EP). The EP subsystem includes the diverter valve.
(a) MT: m˙ 1 + m ˙ 2 =m ˙3
(B1)
21
ACCEPTED MANUSCRIPT ˙ e +Q ˙ l , MT m ˙ 1 h1 + m ˙ 2 LHV = m ˙ 3 h 3+ W
(B2)
˙ l+ E ˙ d) m˙ 1 e 1 + m ˙ 2 e ng =m ˙ 3 e 3 + W˙ e + ( E l,MT
j=3,4,5
(B4)
RI PT
e j =c P,g [ ( T j−T 0 ) −T 0 ln ( T j / T 0 ) ]
(B3)
(b) EP: m ˙ 3 =m ˙4
(B5)
SC
˙ l,EP m ˙ 3 h3= m ˙ 4 h4 + Q
M AN U
m˙ 3 e 3 =m ˙ 4 e 4 + ( E˙ l + E˙ d )l,EP
(c) HRSG: m ˙ 4= m ˙5 m ˙ 6 =m ˙7
(B6) (B7)
(B8) (B9) (B10)
m˙ 4 ( e 4 −e 5 )= m ˙ 6 ( e 6−e 7 ) + ( E˙ l + E˙ d ) HRSG
(B11)
e j =( h j −h 0 )−T 0 ( s j −s 0 )
(B12)
TE D
η HRSG m ˙ 4 c P,g ( T 4 −T 5 ) =m˙ 6 ( h6 −h 7 )
EP
j=5,6
AC C
Since no instrument is installed to measure the steam flow rate entering the absorption chiller (point 6, see Fig. 1), Eq. 10 is considered to estimate the steam mass flow that condenses in the generator jacket.
(d) AC
m ˙ 8 =m ˙9
(B13)
m˙ 6 ( h6 −h 7 )= m ˙ 8 c P,b ( T 9−T 8 ) + Q˙ l,AC
(B14) 22
ACCEPTED MANUSCRIPT m˙ 6 ( e 6 −e 7 ) =m˙ 8 ( e 9 −e 8 ) + ( E˙ l + E˙ d ) AC
j=8,9
(B16)
AC C
EP
TE D
M AN U
SC
RI PT
e j =c P,b [ ( T j −T 0 ) −T 0 ln ( T j /T 0 ) ]
(B15)
23
ACCEPTED MANUSCRIPT
Tables Table 1. Nominal performance data of the chiller (adapted from [18]).
Ambient temperature (°C)
35
Outlet brine temperature (°C)
–5
Cooling capacity (kW)
13.3
COP
0.53
Brine flow for ∆T = 5.5 °C (l/h)
2600
Burner capacity (kW)
25.1
Fuel consumption (m³/h)
2.65
RI PT
Value
SC
Parameter
Table 2. Correction factor for the cooling capacity (adapted from [18]).
–10.0 1.14 1.11 1.06 0.99 0.88 0.72 0.52
M AN U
0, 5, 10, 15 20 25 30 35 40 45
Brine temperature (°C) –7.0 1.15 1.13 1.09 1.03 0.94 0.81 0.63
–5.0 1.15 1.14 1.12 1.07 1.00 0.89 0.75
–2.0 1.16 1.16 1.14 1.10 1.04 0.95 0.81
TE D
Ambient temperature (°C)
0.0 1.17 1.17 1.16 1.14 1.08 1.00 0.87
Table 3. Thermodynamic properties of each material stream. T Stream 24.8 30.5 24.4 305.7 277.4 200.7 160.9 160.9 –4.7 –1.4
AC C
0* 1 2 3 4 5 6 7 8 9
m ˙
e
E˙
(kPa)
(kg/s)
(kJ/kg)
(kW)
– 0.2588 0.0024 0.2612 0.2612 0.2612 0.0092 0.0092 0.7702 0.7702
– 0 49152 86.7 72.3 38.5 753.2 102.2 5.7 4.5
– 0 119.5 22.7 18.9 10.1 6.9 0.9 4.4 3.4
EP
(°C)
P
101.3 101.3 195.1 101.3 101.3 101.3 633.7 633.7 301.3 301.3
* Environment conditions.
24
ACCEPTED MANUSCRIPT
Stream Natural gas Power Heat (steam @ 5.3 bar) Cooling (brine @ –5 °C)
energy
exergy
(kW) 114.4 26.0 19.0 9.2
(kW) 119.5 26.0 6.9 0.9
RI PT
Table 4. Energy and exergy of each energy stream.
Table 5. Performance parameters of each plant subsystem. ε
COP
(–) 0.227 – 0.920 – 0.393
(–) 0.218 – 0.674 0.158 0.275
(–) – – – 0.440 –
E˙ d
yd
SC
MT EP HRSG AC Plant
η
(kW) 70.9 3.8 2.9 5.9 83.5
(–) 0.593 0.032 0.024 0.042 0.699
M AN U
Subsystem
Table A1. Cost flow rate and specific exergy cost of each energy stream (fuel and products). Fuel/ Product
Natural gas Power Cooling (brine @ –4.7°C)
F P P
C˙
c
(USD/h) 7.01 10.8 1.91
(USD/kWh) 0.059 0.416 2.026
TE D
Stream
E˙ d
Subsystem
EP
Table A2. Exergoeconomic parameters of each plant subsystem.
(–)
AC C
(kW)
yd
MT EP HRSG AC
70.9 3.8 2.9 5.9
0.593 0.032 0.024 0.042
cf
cp
Z˙
(USD/kWh) (USD/kWh) (USD/h) 0.06 0.06 0.07 0.15
0.24 0.07 0.15 1.65
4.77 0 0.26 0.67
C˙ d
r
f
(USD/h)
(–)
(–)
4.16 0.22 0.20 0.74
3.13 0.20 1.10 10.2
0.534 0 0.562 0.475
25
ACCEPTED MANUSCRIPT
Figures
RI PT
Figure 1. Scheme of the small scale cogeneration plant.
Figure 2. Chiller generator: (a) original design; (b) modified for cogeneration.
SC
Figure 3. Variation of the steam pressure with power output.
M AN U
Figure 4. Variation of the cooling capacity with steam pressure.
Figure 5. Variation of the cooling capacity with chiller outlet temperature.
TE D
Figure 6. Variation of the coefficient of performance with steam pressure.
Figure 7. Variation of the coefficient of performance with chiller outlet temperature.
EP
Figure 8. Variation of the chiller exergetic efficiency with steam pressure.
AC C
Figure 9. Variation of the chiller exergetic efficiency with cooling temperature.
26
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S1359-4311(13)00312-8
DOI:
10.1016/j.applthermaleng.2013.04.041
Reference:
ATE 4775
To appear in:
Applied Thermal Engineering
Received Date: 19 November 2012 Accepted Date: 25 April 2013
Please cite this article as: E. Bazzo, A. Nacif de Carvalho, J.A. Matelli , Experimental results and thermodynamic analysis of a natural gas small scale cogeneration plant for power and refrigeration purposes, Applied Thermal Engineering (2013), doi: 10.1016/j.applthermaleng.2013.04.041. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights of the manuscript entitled
EXPERIMENTAL RESULTS AND THERMODYNAMIC ANALYSIS OF A NATURAL GAS SMALL SCALE COGENERATION PLANT FOR POWER AND REFRIGERATION PURPOSES
RI PT
by Edson Bazzo, Alvaro N. Carvalho and José A. Matelli
A small scale cogeneration plant for power and refrigeration is proposed and analyzed.
•
The plant is based on a microturbine and a modified absorption chiller;
•
The plant is analysed based on 1st and 2nd laws of thermodynamics.
•
Experimental results are found for different power and refrigeration conditions.
•
The plant proved to be technically feasible.
AC C
EP
TE D
M AN U
SC
•
ACCEPTED MANUSCRIPT
RI PT
A manuscript entitled:
M AN U
by:
SC
EXPERIMENTAL RESULTS AND THERMODYNAMIC ANALYSIS OF A NATURAL GAS SMALL SCALE COGENERATION PLANT FOR POWER AND REFRIGERATION PURPOSES
Edson Bazzo1
Alvaro Nacif de Carvalho1 José Alexandre Matelli2†
TE D
† Author for correspondence [email protected] Phone: +55 12 3123-2839 fax: +55 12 3123-2835
AC C
EP
1. UFSC – Federal University of Santa Catarina Department of Mechanical Engineering 88040-900 Florianópolis, SC – Brazil 2. UNESP – Univ Estadual Paulista Department of Energy 12516-410 Guaratinguetá, SP – Brazil
To be submitted to the Applied Thermal Engineering
Guaratinguetá, winter 2012.
1
ACCEPTED MANUSCRIPT
Abstract. In this work, experimental results are reported for a small scale cogeneration plant for power and refrigeration purposes. The plant includes a natural gas microturbine and an ammoniawater absorption chiller fired by steam. The system was tested under different turbine loads, steam
RI PT
pressures and chiller outlet temperatures. An evaluation based on the 1st and 2nd Laws of Thermodynamics was also performed. For the ambient temperature around 24°C and microturbine at full load, the plant is able to provide 19 kW of saturated steam at 5.3 bar (161ºC), corresponding to 9.2 kW of refrigeration at –5°C (COP = 0.44). From a 2nd law point-of-view, it was found that
SC
there is an optimal chiller outlet temperature that maximizes the chiller exergetic efficiency. As expected, the microturbine presented the highest irreversibilities, followed by the absorption chiller
M AN U
and the HRSG. In order to reduce the plant exergy destruction, it is recommended a new design for the HRSG and a new insulation for the exhaust pipe.
AC C
EP
TE D
Keywords. Cogeneration, Microturbine, Absorption chiller, Natural gas.
2
ACCEPTED MANUSCRIPT
Nomenclature c
specific exergy cost [USD/kWh]
Subscripts 0
reference state
coefficient of peformance [–]
b
brine
C˙
exergy cost rate [USD/h]
d
destruction
e
specific exergy [kJ/kg]
e
electric
E˙
exergy rate [kW]
f
fuel
f
exergoeconomic factor [–]
g
exhaust gases
h
specific enthalpy [kJ/kg]
i
inlet stream
LHV
lower heat value [kJ/kg]
j
j-th stream
m ˙
mass flow [kg/s]
k
k-th plant component
P
pressure [kPa]
l
lost
˙ Q
heat transfer rate [kW]
r
relative cost difference [–]
s
specific entropy [kJ/kgK]
T
temperature [°C]
W˙
work transfer rate [kW] exergy destruction ratio [–]
Z˙
purchase cost rate [USD/h]
Greek letters
TE D
y
energy-based efficiency [–]
ε
exergy-based efficiency [–]
natural gas
o
outlet stream
p
product
q
heat related
s
steam
w
work related
Acronyms AC
absorption chiller
EP
exhaust pipe
HRSG heat recovery steam generator MT
microturbine
AC C
EP
η
SC
ng
M AN U
COP
RI PT
specific heat [kJ/kgK]
cP
3
ACCEPTED MANUSCRIPT
1. Introduction
The use of waste heat as energy source for absorption chillers can be an alternative to decrease
RI PT
operating costs of either air conditioning or refrigeration processes. Usually, the waste heat comes from gas turbines, reciprocating engines or from some other industrial processes. The use of
cogeneration systems based on combined cooling, heating and power plants (CCHP) in the industry [1, 2] and in commercial buildings [3, 4] was investigated and it is well documented in the
SC
literature.
Recent results related to experimental and theoretical investigations of small scale cogeneration
M AN U
plants based on microturbines and absorption chillers are reported [5–10], demonstrating the technical feasibility and the potential to achieve high overall performance. Aiming to assure the feasibility of these systems, several heat recovery technologies for refrigeration are recommended for different levels of temperature [11]. The influence of the heat source temperature on COP and
TE D
cooling output of H2O-LiBr and H2O-NH3 absorption chillers are also investigated [12–16]. Recently, Rossa and Bazzo [17] reported a thermodynamic first and second law analysis of a small scale cogeneration plant for power and air-conditioning. The plant consists of a 28 kWe
EP
natural gas microturbine, a thermosyphon heat exchanger (the heat from the microturbine exhaust gases is recovered to generate hot water) and a H2O-NH3 absorption chiller using hot water as heat
AC C
source.
The objective of the present paper is to report the experimental data of an upgraded version of the plant studied by Rossa and Bazzo [17]. The new plant is conceived for refrigeration purposes and consists of a 28 kWe natural gas microturbine, a HRSG (the heat from the microturbine exhaust gases is recovered to generate saturated steam) and a H2O-NH3 absorption chiller using saturated steam as heat source. Instruments for pressure, mass flow rates and temperature were placed in key points of the plant. The tests were carried out for different levels of power output power, steam pressure and chiller output temperature, with the respective experimental results reported in this paper. In order to identify possible improvements in the plant performance, a thermodynamic 4
ACCEPTED MANUSCRIPT
analysis is held. Some changes are proposed to reduce the exergy destruction and to increase the overall efficiency.
RI PT
2. Small scale cogeneration plant description
The schematic diagram of the small scale cogeneration plant is shown in Fig. 1. A set of six
cylinders store and supply natural gas to the microturbine. The heat from the turbine exhaust gases
SC
is recovered in the HRSG, where steam is generated in order to feed the H2O-NH3 chiller. The cooling output from the chiller is a brine solution of monoethyleneglycol, which is stored in a
M AN U
proper tank. The tank has electrical resistors to emulate a thermal load. In addition, volumetric flow rate devices as well as temperature and pressure sensors are all connected to a data acquisition system.
TE D
Figure 1. Scheme of the small scale cogeneration plant.
The prime mover is a Capstone C30 LP microturbine. It includes a compressor, a regenerator, a
EP
combustion chamber, a turbine and a generator. The rotating components are all assembled over a single shaft supported by air bearings and can reach angular velocities as high as 96000 rpm. There
AC C
is no reduction gearbox between the generator shaft and the gas turbine shaft, so that the generator power output is electronically treated in order to provide proper values of AC voltage and frequency. Operating at ISO condition (15°C and 101.3 kPa), the microturbine generates 28 kW of power and 84 kW of recoverable heat from 0.31 kg/s of exhaust gases at 275 °C, with an electrical efficiency equal to 0.25 [18]. The HRSG consists of a diverter valve, a cross-flow heat exchanger and a steam drum. The residual heat from the microturbine is recovered into the HRSG to generate saturated steam for the chiller. For security reasons, the HRSG is designed for a maximum pressure of 6 bar. The heat input to the HRSG is automatically controlled by an on-off diverter valve, which acts according the steam 5
ACCEPTED MANUSCRIPT
pressure signal. Thus, the exhaust gases are either diverted directly to the atmosphere to avoid over pressure or flows through the heat exchanger to keep the pressure between specified values. The absorption chiller is a Robur ACF60-00 LB designed for refrigeration, with cooling
RI PT
capacity of 13.3 kW and output temperature of –5°C. Additional performance data of the chiller are presented in Tab. 1. These figures are rated by the manufacturer at an ambient temperature of 35°C [19]. The manufacturer provides a correction factor (taken into account in this work) to predict the chiller performance at different ambient temperatures and different cooling temperatures, as shown
SC
in Tab. 2 [19].
M AN U
Table 1. Chiller performance data (adapted from [19]).
Table 2. Correction factor for the cooling capacity (adapted from [19]).
The chiller is a single-block unit based on a H2O-NH3 solution and single effect cycle with an
TE D
air-cooled condenser. It is designed to work with direct natural gas firing through a burner. The heat from the combustion is transferred to the chiller generator. Since the heat is transferred mainly by convection, the generator has fins in its external surface (Fig. 2a). In order to make the chiller
EP
suitable for cogeneration, the burner was removed and a properly designed jacket was installed. The generator is now fully enveloped by the jacket, so that the steam enters the jacket, flows around the
AC C
external generator fins and leaves the jacket as condensate (Fig. 2b). When a steady-state condition is established during a particular test, the steam mass flow remains constant. However, the steam mass flow may vary from one steady-state test to another. Since no instrument is installed to measure the steam flow rate entering the absorption chiller, an energy balance in the HRSG is considered to estimate the steam mass flow that condenses in the generator jacket. Other important modification was made in the chiller control system, originally designed to control the heat input in the chiller generator according the cooling outlet temperature. Basically, this system was shutdown and replaced by a customized one. Since the outlet cooling temperature and the steam pressure are
6
ACCEPTED MANUSCRIPT
related, the steam pressure signal is used to control the HRSG diverter valve and, consequently, the heat input to the chiller generator. The chiller cooling output is a brine solution of water-
RI PT
monoethyleneglycol (33% in mass) able to reach temperatures below –10°C without freezing.
Figure 2. Chiller generator: (a) original design; (b) modified for cogeneration.
SC
The brine circulates continually between the tank and the chiller. Four electrical resistors of 4.8 kW each are immersed in the tank in order to emulate a thermal load and to allow steady state plant
M AN U
operation.
A particularity of this plant is that the microturbine is placed indoors and all other components are placed outdoors. Thus, the microturbine and the chiller can be subject to slightly different
3. Methodology
TE D
ambient temperatures.
EP
3.1. Plant experimental tests
For the power output, the measurements of power, voltage and current can be taken directly
AC C
from the microturbine control system. For the cooling output, measurements of temperatures, pressures and flow rates in strategic points of the plant are used to obtain thermodynamic properties for further evaluation. These measurements are available from the data acquisition system. In order to evaluate the influence of the steam pressure on the chiller performance, the microturbine was set to different power outputs. For each power output there is a corresponding exhaust gases flow and temperature. Hence, for each power output there is a corresponding steam pressure in the HRSG. Since the tests were carried out during several days, the weather forecast was checked in order to assure some repeatability in the ambient temperature.
7
ACCEPTED MANUSCRIPT
The tests were performed at steady state condition for different power outputs: 16, 18, 20, 22, 24 and 26 kW. When the pressure was stable in the HRSG, the absorption chiller worked with different cooling outputs by setting the brine temperature to 0, –2, –5, –7 and –9.5 °C (point 8, see Fig. 1).
RI PT
During the tests, the ambient temperature ranged between 22.1°C and 24.5°C. Although the nominal cooling output and COP were rated by the chiller manufacturer at 35°C, an available correction factor (Tab. 2) was used for different ambient temperatures.
SC
3.2. Thermodynamic analysis
M AN U
A thermodynamic analysis is performed in order to evaluate the plant performance and consolidate the results from the experiments. The analysis considers four subsystems: the microturbine (MT), the heat recovery steam generator (HRSG), the absorption chiller (AC) and the exhaust gases pipe from the microturbine to the steam generator (EP). The EP subsystem includes
TE D
the diverter valve. Since the exhaust gases pipe is relatively long, the EP subsystem is considered to evaluate the exhaust gas losses between the microturbine outlet and the steam generator inlet. This analysis focuses on the identification of the irreversibilities. The cost contribution of each
EP
subsystem and an exergoeconomic analysis are found in the Appendix A. The following assumptions are considered in the analysis:
• • •
Steady state condition;
AC C
•
Exhaust gases properties as air properties; Ideal gas model;
Negligible kinetic and potential energy changes;
•
Saturated steam (x=1) at HRSG outlet;
•
Saturated liquid (x=0) at HRSG inlet;
•
Negligible liquid pumping power consumption;
•
Negligible pressure drop through the HRSG;
8
ACCEPTED MANUSCRIPT •
For simplicity, the HRSG efficiency is constant and equal to 0.92.
Mass, energy and exergy balances are considered for each subsystem, taking into account heat
RI PT
transfer, work interaction and the inlet/outlet material streams, as shown in Eqs. 1–3. The detailed set of equations is shown in Appendix B. The thermodynamic properties are obtained from the Engineering Equation Solver library. All the equations refers to the points depicted in Fig. 1.
SC
∑ m˙ i =∑ m˙ o
E˙ q −W˙ =∑ m ˙ o e o −∑ m˙ i ei + E˙ D
M AN U
˙ W˙ =∑ m˙ o ho −∑ m˙ i hi Q−
(1) (2) (3)
Considering the absence of entropy data for the brine solution of monoethyleneglycol, the corresponding exergy was estimated by Eq. B16. The First Law-parameters described in Eqs. 4–7
η e=W˙ e / ( m ˙ 2 LHV )
Q˙ s= m ˙ 6 ( h 6−h7 )
EP
η ov=( W˙ e+ Q˙ s ) / ( m˙ 2 LHV )
TE D
are considered to evaluate the performance of the plant:
AC C
COP= [ m ˙ 8 c P,b ( T 9 −T 8 ) ] / [ m ˙ 4 c P,g ( T 4 −T 5 ) ]
(4) (5) (6) (7)
Based on the natural gas composition informed by the local gas utility, the lower heating value and the exergy of the fuel are calculated as 47308 kJ/kg and 49437 kJ/kg, respectively. The exergy of the fuel components is given in Kotas [20]. The overall efficiency in Eq. 5 is stated considering the steam input in the chiller instead of its cooling output. In this way, it is possible to evaluate the maximum amount of fuel energy that is effectively converted into power or steam for comparison with other cogeneration systems. So, the absorption chiller is considered as a subsystem that uses
9
ACCEPTED MANUSCRIPT
the available steam as energy source. Thus, the COP (Eq. 7) is the parameter that completes the plant evaluation. The parameters shown in Eqs 8–12 are considered for evaluation based on the Second Law, on
RI PT
which the irreversibilities and exergy destructions are identified. For the microtubine in particular, both exergy destruction and exery lost were considered in the definition of its exergy efficiency. The exergy destruction ratio of a subsystem k, as shown in the Eq. 12, represents the contribution of
SC
exergy destruction of a subsystem related to the total exergy invested in the system.
ε HRSG =[ m ˙ 6 ( e 6− e 7 ) ] / [ m ˙ 4 ( e 4 −e 5 ) ] ε AC=[ m ˙ 8 ( e 9−e8 ) ] / [ m ˙ 6 ( e 6−e 7 ) ]
˙ e +m ε ov =[ W ˙ 6 ( e 6 −e 7 ) ] / ( m ˙ 2 e ng )
(8) (9) (10) (11) (12)
EP
4. Results
TE D
y d,k = E˙ k / ( m ˙ 2 e ng )
M AN U
˙ e / ( m˙ 2 e ng ) ε MT =W
AC C
4.1. Experimental tests results
The steam pressure obtained in the HRSG for different microturbine power output is shown in Fig. 3. As expected, the higher the power output, the higher the steam pressure obtained in the HRSG. It is important to remind that a steady state condition was assured for each power output. The linear behavior presented in Fig. 3 is also expected, since the microturbine exhaust gases heat increases linearly with power output. In addition, the steam generation is a thermodynamic process at constant volume. In such a process, the heat transferred from the turbine exhaust gases increases
10
ACCEPTED MANUSCRIPT
the steam enthalpy. Since the pressure variation is approximately linear with the steam enthalpy, the steam pressure increases linearly with turbine power output.
RI PT
Figure 3. Variation of the steam pressure with power.
For different brine outlet temperatures, the cooling capacity variation with steam pressure is
shown in Figure 4. Again, a linear trend can be observed for the different brine temperatures. It is
SC
explained as follows: the ammonia-water evaporation in the chiller generator is a process at constant volume. In addition, the mass balance in the generator requires that the ammonia mass
M AN U
fraction remains constant. Consider that the heat transferred to the generator increases and both the volume and the ammonia mass fraction remain constant. In this case, the amount of ammonia-water vaporized increases linearly with the heat transferred, resulting in a higher refrigerant mass flow in the evaporator and, consequently, a higher cooling capacity. Since the heat transferred to the
TE D
generator is directly related to the steam pressure, it follows that the cooling capacity increases linearly with the steam pressure.
EP
Figure 4. Variation of the cooling capacity with steam pressure.
AC C
For different steam pressures, the cooling capacity as a function of the brine outlet temperature is shown in Figure 5. As well known in refrigeration systems, the cooling capacity increases linearly with the evaporation temperature, which is also linearly related to the brine temperature in this case. The direct firing points shown in Fig. 5 are calculated values from the manufacturer documentation [19] rather than measured data. These points are presented in order to provide a comparison between the cooling capacity of the modified generator and its original design. At full load (26 kWe), nominal chiller outlet brine temperature (–5°C) and ambient temperature around 24 °C, the absorption chiller provided 9.2 kW of cooling capacity. Considering the correction factor for
11
ACCEPTED MANUSCRIPT
25 °C presented in Tab. 2, the direct firing system provides 14.9 kW of cooling capacity for the same brine temperature. Thus, a reduction of 38% in the cooling capacity is observed.
RI PT
Figure 5. Variation of the cooling capacity with chiller outlet temperature.
Figure 6 shows the variation of the coefficient of performance with steam pressure for different cooling temperatures. The observed trend is that the COP remains approximately constant for the
SC
pressure range tested. This result is in accordance with the results from the work of Fernandéz-
Seara and Vásquez [15], on which the authors reported no significant changes in the COP when the
M AN U
heat source temperature of a single effect ammonia/water absorption chiller ranged from 138°C (2.4 bar) to 161°C (5.3 bar).
Figure 6. Variation of the coefficient of performance with steam pressure.
Figure 7 shows the COP variation with chiller outlet temperature. Again, it is well known that
TE D
the COP increases linearly with evaporation temperature. When fed by steam, the chiller presents a COP reduction relative to its original design. At full load (26 kWe), nominal refrigeration temperature (–5°C) and ambient temperature around 24 °C, a COP of 0.44 was found. That means
EP
25% less if compared to the nominal value of 0.59 for the direct firing system. It can be also seen that the performance of the modified chiller is lower than the direct-firing for all tests. It can be
AC C
explained because the chiller generator is originally designed for both convection and radiation heat transfer from a natural gas flame, as long as operating with steam, the heat transfer is mainly from changing phase convection. Additionally, the generator fins impose a high heat transfer resistance when the convection heat transfer coefficient is as high as in changing phase convection. It is important to remind that the COP as presented in Figs 6 and 7 is calculated based on the heat recovered from the exhaust gases.
Figure 7. Variation of the coefficient of performance with chiller outlet temperature.
12
ACCEPTED MANUSCRIPT
Experimental results considering the 2nd law are also presented. Figure 8 shows that the chiller exergetic efficiency decreases with the steam pressure. This is expected, because the temperature
RI PT
difference in any heat transfer process causes entropy generation, unless the temperature difference tends to zero. A higher steam pressure means a higher temperature difference between the steam and the ammonia-water solution in the generator, so that the entropy generation is higher at high
SC
steam pressures. Thus, the exergy efficiency of the chiller decreases with steam pressure.
M AN U
Figure 8. Variation of the chiller exergetic efficiency with steam pressure.
Figure 9 presents the variation of the chiller exergetic efficiency with cooling temperature for different steam pressures. The results show that, for any steam pressure, the chiller exergetic
TE D
efficiency reaches a maximum value for the refrigeration temperatures around –7°C.
EP
Figure 9. Variation of the chiller exergetic efficiency with cooling temperature.
AC C
4.2. Thermodynamic analysis results
The thermodynamic analysis is held considering the experimental data when the microturbine operated at full load (26 kWe) and ambient temperature equal to 24.4 °C and the chiller at nominal cooling temperature (–5 °C) and ambient temperature equal to 24.8 °C. This slightly difference in the ambient temperature is observed because the microturbine is placed indoors and all other components are placed outdoors. The thermodynamic properties (temperature, pressure and mass flow rate and exergy) related to the material streams are all shown in Tab. 3. The material streams are labeled as shown in Fig. 1.
13
ACCEPTED MANUSCRIPT
Table 3. Thermodynamic properties of each material stream.
RI PT
The energy and exergy related to the energy streams are shown in Tab. 4. The first and second law parameters of performance for each plant subsystem are shown in Tab. 5.
Table 4. Energy and exergy of each energy stream.
SC
Table 5. Performance parameters of each plant subsystem.
M AN U
By far the microturbine presents the higher exergy destruction of the plant, which is expected due to the internal combustion process. Except for pre-heating the reactants or adjusting the air-fuel ratio, nothing can be done to minimize the exergy destruction of a combustion process. Since the microturbine presents an air preheater and its air-fuel ratio is preset, no further actions can be taken
considered.
TE D
in this case. Thus, the replacement of the current microturbine for a more efficient one should be
Regarding the HRSG, the sources of exergy destruction are friction and, most significantly,
EP
stream-to-steam heat transfer. Thus, the main action that could be taken is to optimize the HRSG heat exchange surface in order to obtain pinch point values (∆Tpp = T5 – T6) within practical
AC C
recommended values (10–25°C). A design that minimizes the gas stream pressure drop contributes to reduce the exergy destruction related to the friction. It can be also observed that the HRSG outlet temperature (point 5) indicates that there is still room for heat recovering, since it is well known that this temperature can be as low as 110 °C. The poor insulation in the exhaust pipe between the microturbine and HRSG causes an exergy destruction even higher than the exergy destroyed in the HRSG itself. Thus, a new insulation for the exhaust pipe is recommended. The insulation thickness of a non-isothermal pipe that minimizes the exergy destruction can be determined according the methodology proposed by Bejan et al. [21]. Due to the significant temperature difference between the steam and the ammonia-water solution in 14
ACCEPTED MANUSCRIPT
the generator, there is also exergy destruction in the chiller. The recommendation in this case is a new generator design that maximizes the convection heat transfer from the condensing steam. The results from the thermodynamic analysis are complemented by the results from the
RI PT
exergoeconomic analysis presented in the Appendix A.
5. Conclusion
SC
An experimental study related to a small scale cogeneration plant, consisting of a microturbine, a heat recovery steam generator and an ammonia-water absorption chiller was presented. The chiller
M AN U
was modified in order to use saturated steam as heat source instead the original natural gas directfiring system. The steam is generated from the microturbine exhaust gases heat. The plant proved to be technically feasible for power generation and refrigeration for freezing purposes. For different microturbine power output, the chiller performance was evaluated for different cooling
TE D
temperatures. It is important to emphasize that the tests were carried out during several days. Although weather forecast was checked in order to assure repeatability, some variation in the ambient temperature was inevitability observed. At full load and ambient temperature around 24°C,
EP
the cogeneration plant provided 26 kW of power and 19 kW of saturated steam at 5.3 bar (161ºC), so that 9.2 kW of cooling capacity was generated at –5 °C. The COP was found as 0.44 and the
AC C
overall efficiency was found as 39.4%. Thus, the chiller performance operating in cogeneration is 25% less than the nominal value of 0.59 for the direct firing system. The chiller modification explains this result, because the chiller generator is originally designed for both convection and radiation heat transfer from a natural gas flame, as long as operating with steam, the heat transfer is mainly from changing phase convection. Additionally, the generator fins impose a high heat transfer resistance when the convection heat transfer coefficient is as high as in changing phase convection. However, the HRSG outlet temperature indicates that there is room for heat recovery improvement. Higher steam pressures lead to a substantial increase in the chiller output, although no significant changes were observed in the COP. Both cooling output and COP decreased when 15
ACCEPTED MANUSCRIPT
using the residual heat system with steam, comparing to the direct firing system values available in the manufacturer’s catalogue. From a 2nd law point-of-view, it was found that there is an optimal chiller outlet temperature that maximizes the chiller exergetic efficiency. As expected, the
RI PT
microturbine presented the highest irreversibilities, followed by the absorption chiller and the HRSG. In order to reduce the plant exergy destruction, it is recommended a new design for the HRSG and a new insulation for the exhaust pipe.
SC
Acknowledgements
M AN U
The authors appreciate the FINEP (Research and Projects Financing), the CNPq (National Council for Scientific and Technological Development) and the companies Petrobras, TBG and SCGás, for financial support granted to this research. The authors are also thankful to the Labsolda/UFSC for building the customized control system for the chiller. Part of this research was
AC C
EP
TE D
developed under the cooperation agreement between UFSC and UNESP.
16
ACCEPTED MANUSCRIPT
References
[1] L. Trygg, S. Amiri, European perspective on absorption cooling in a combined heat and power
1337.
RI PT
system – A case study of energy utility and industries in Sweden, Appl. Energy, 84 (2007) 1319–
[2] I. Svensson, B. Moshfegh, System analysis in a European perspective of new industrial cooling supply in a CHP system, Appl. Energy, 88 (2011) 5164–5172.
SC
[3] G.G. Maidment, X. Zhao, S.B. Riffat, G. Prosser, Application of combined heat-and-power and absorption cooling in a supermarket, Appl. Energy, 63 (1999) 169–190
M AN U
[4] G.G. Maidment, R.M. Tozer, Combined cooling heat and power in supermarkets, Appl. Therm. Eng. 22 (2002) 653–665.
[5] S. Tassou, I. Chaer, N. Sugiartha, Y. Ge, D. Marriott, Application of tri-generation systems to the food retail industry, Energy Convers. Manag. 48 (2007) 2988–2995.
TE D
[6] L. Colombo, F. Armanasco, O. Perego, Experimentation on a cogenerative system based on a microturbine, Appl. Therm. Eng. 27 (2007) 705–711. [7] A. Zaltash, A., Petrov, D. Rizy, S. Labinov, E. Vineyard, R. Linkous, Laboratory R&D on
EP
integrated energy systems (IES), Appl. Therm. Eng. 26 (2006) 28–35. [8] M. Medrano, J. Mauzey, V. McDonell, S. Samuelsen, D. Boer, Theoretical Analysis of a Novel
AC C
Integrated Energy System Formed by a Microturbine and an Exhaust Fired Single-Double Effect Absorption Chiller, Int. J. Thermodyn. 9 (2006) 29–36. [9] J. Ho, K. Chua, S. Chou, Performance study of a microturbine system for cogeneration application, Renew. Energy, 29 (2004) 1121–1133. [10] M.S. Rocha, R. Andreos, J.R. Simões-Moreira, Performance tests of two small trigeneration pilot plants, Appl. Therm. Eng. 41 (2012) 84–91. [11] D. Wu, R. Wang, Combined cooling, heating and power: A review, Energy Combust. Sci. 32 (2007) 459–495.
17
ACCEPTED MANUSCRIPT
[12] G. Grossman, Solar-powered systems for cooling, dehumidification and air-conditioning, Solar Energy, 72 (2002) 53–62. [13] M. Izquierdo, E. Palacios, J. Marcos, D. Mendoza, Air cooling by a solar absorption system:
RI PT
2008 summer experimental results”, in: Proceedings of ECOS 2009, Aug. 31 - Sept. 03, 2009, Foz do Iguaçu, Brazil.
[14] A. Keçeciler, H. Acar, A. Dogan, Thermodynamic analysis of the absorption refrigeration
system with geothermal energy: an experimental study, Energy Convers. Manag. 41 (2000) 37–48.
SC
[15] J. Fernández-Seara, M. Vázquez,Study and control of the optimal generation temperature in NH3-H2O absorption refrigeration systems, Appl. Therm. Eng. 21 (2001) 343–357.
Int. J. Refrig. 27 (2004) 10–16.
M AN U
[16] I. Horuz, T. Callander, Experimental investigation of a vapor absorption refrigeration system,
[17] J. A. Rossa, E. Bazzo, Thermodynamic modeling for an ammonia-water absorption system associated to a microturbine, Int. J. Thermodyn. 12 (2009) 38–43.
March 27, 2013.
TE D
[18] Capstone C30 LP. Product catalogue. Available at http://www.capstoneturbine.com, access
[19] Robur ACF 60-00 LB. Installation, use and maintenance instructions (in Italian). Available at
EP
http://www.robur.it, access March 27, 2013.
[20] T. Kotas, The exergy method of thermal plant analysis, Krieger Publishing Company, Malabar,
AC C
1995.
[21] A. Bejan, G. Tsatsaronis, M. Moran, Thermal Design and Optimization, Jonh Wiley & Sons, New York, 1996.
18
ACCEPTED MANUSCRIPT
Appendix A: Exergoeconomic analysis
An exergoeconomic analysis is also performed for understanding the cost formation process and
RI PT
the costs flow in the system, so that potential improvements are identified in order to minimize the final cost of the products. The exergoeconomic model consists of a balance of these costs. Actual purchase and installation costs of the plant in the laboratory were taken into account; fuel and power prices are given by the respective local utilities. Additional expenses were considered as 50% of the
SC
total purchase equipment costs. All costs were converted to US dollars with an exchange ratio of 1.71 R$/US$ at the time this paper was written. The purchase costs of the equipments were
M AN U
levelized based on 20 years of operation, 7446 hours of operation per year, 12% of interest loan and 5% for operation and maintenance costs per year. The inflation rate was not considered. The equations presented in this section were stated in accordance with methodology presented in Bejan et al. [21]. The cost rate for a stream j and the cost balance for a component k are given as the
TE D
following:
C˙ j =c j E˙
j
C˙ w,k + ∑ C˙ o,k = C˙ q,k + ∑ C˙ i,k + Z˙ k o
i
EP
The cost related to the lost and destroyed exergy in a component k depends on the stream that
AC C
carry the lost and destroyed exergy out of the component: C˙ d,k =c l,k ( E˙ d + E˙ l )k
If power and brine are considered as plant final products, the cost flow and the specific cost of the brine are respectively given as follows: ˙ ) + C˙ /2 C˙ b =( C˙ 8 −C 9 5 ˙ c b =C˙ b / W b
The relative cost difference (Eq. A1) measures the relative increase in the specific cost between fuel and product of each subsystem or component. The exergoeconomic factor (Eq. A2) expresses
19
ACCEPTED MANUSCRIPT
the weight of the sources on the cost aggregation, whether it is greater for the subsystem purchase price or for exergy destruction and loss. c p,k − c f,k c f,k
f k=
(A1)
Z˙ k Z˙ k + (C˙ d + C˙ l )
RI PT
r k=
(A2)
SC
The exergoeconomic model is evaluated considering the experimental data when the
microturbine operated at full load (26 kWe) and the chiller at nominal cooling temperature (–5 °C).
M AN U
The cost flow rate and specific exergy cost of the fuel and products are presented in Table A1. The steam cost is applicable only in case of the cogeneration plant generating power and steam as final products. According to the local power utility, the power price is 0.571 US$/kWh for the peak period and 0.084 US$/kWh for the off-peak period. These figures show that the power generated by
TE D
the cogeneration plant is far more expensive than the current market off-peak power prices. On the other hand, it is competitive when compared to the peak power price. Table A2 shows the exergoeconomic parameters regarding the plant subsystems.
AC C
EP
Table A1. Cost flow rate and specific exergy cost of each energy stream (fuel and products).
Table A2. Exergoeconomic parameters of each plant subsystem.
As expected, the microturbine is the subsystem that most contribute for cost aggregation in the system. The microturbine exergoeconomic factor (f) found here is in accordance with the figures found in the literature, as those found in Bejan et al. [20]. The cost is quite well distributed between investment cost and exergy destruction cost. However, the high relative cost difference (r) indicates that cost contribution of the microturbine in the overall system is high.
20
ACCEPTED MANUSCRIPT
The absorption chiller presented the highest relative cost difference due to the low exergy related to its products (streams 8 and 9). Equipments for refrigeration purposes will always aggregate a high cost per exergy unit to the streams. In order to reduce this factor, the chiller should
RI PT
operate at high exergy efficiencies. In the present work, the experimental results show that this operation point corresponds to an outlet brine temperature of –7°C (Fig. 9). The exergoeconomic factor shows that purchase costs and exergy destruction costs are quite balanced. However, the lack of information for cooling equipments in the literature does not allow a more precise analysis.
SC
As the microturbine, the HRSG presented a typical value of exergoeconomic factor. The relative cost factor was found to be high. In this case, it is possible to enhance the exergetic efficiency by
M AN U
using a HRSG with lower pinch point. The exergy destruction in the exhaust pipe causes an increase of 20% on the specific exergy cost. The poor insulation between the microturbine outlet and the HRSG inlet causes an additional cost of 0.38US$/h to the fuel of this subsystem to cover the exergy destruction. A review of the insulation is recommended. In general, it was observed that
efficiency/capacity ratios.
TE D
small scale equipments present a high relative cost factor, due to the high cost/capacity and low
EP
Appendix B: Detailed set of equations of the thermodynamic analysis
AC C
Mass, energy and exergy balances are considered for each subsystem, taking into account heat transfer, work interaction and the inlet/outlet material streams. The subsistems are the microturbine (MT), the heat recovery steam generator (HRSG), the absorption chiller (AC) and the exhaust gases pipe from the microturbine to the steam generator (EP). The EP subsystem includes the diverter valve.
(a) MT: m˙ 1 + m ˙ 2 =m ˙3
(B1)
21
ACCEPTED MANUSCRIPT ˙ e +Q ˙ l , MT m ˙ 1 h1 + m ˙ 2 LHV = m ˙ 3 h 3+ W
(B2)
˙ l+ E ˙ d) m˙ 1 e 1 + m ˙ 2 e ng =m ˙ 3 e 3 + W˙ e + ( E l,MT
j=3,4,5
(B4)
RI PT
e j =c P,g [ ( T j−T 0 ) −T 0 ln ( T j / T 0 ) ]
(B3)
(b) EP: m ˙ 3 =m ˙4
(B5)
SC
˙ l,EP m ˙ 3 h3= m ˙ 4 h4 + Q
M AN U
m˙ 3 e 3 =m ˙ 4 e 4 + ( E˙ l + E˙ d )l,EP
(c) HRSG: m ˙ 4= m ˙5 m ˙ 6 =m ˙7
(B6) (B7)
(B8) (B9) (B10)
m˙ 4 ( e 4 −e 5 )= m ˙ 6 ( e 6−e 7 ) + ( E˙ l + E˙ d ) HRSG
(B11)
e j =( h j −h 0 )−T 0 ( s j −s 0 )
(B12)
TE D
η HRSG m ˙ 4 c P,g ( T 4 −T 5 ) =m˙ 6 ( h6 −h 7 )
EP
j=5,6
AC C
Since no instrument is installed to measure the steam flow rate entering the absorption chiller (point 6, see Fig. 1), Eq. 10 is considered to estimate the steam mass flow that condenses in the generator jacket.
(d) AC
m ˙ 8 =m ˙9
(B13)
m˙ 6 ( h6 −h 7 )= m ˙ 8 c P,b ( T 9−T 8 ) + Q˙ l,AC
(B14) 22
ACCEPTED MANUSCRIPT m˙ 6 ( e 6 −e 7 ) =m˙ 8 ( e 9 −e 8 ) + ( E˙ l + E˙ d ) AC
j=8,9
(B16)
AC C
EP
TE D
M AN U
SC
RI PT
e j =c P,b [ ( T j −T 0 ) −T 0 ln ( T j /T 0 ) ]
(B15)
23
ACCEPTED MANUSCRIPT
Tables Table 1. Nominal performance data of the chiller (adapted from [18]).
Ambient temperature (°C)
35
Outlet brine temperature (°C)
–5
Cooling capacity (kW)
13.3
COP
0.53
Brine flow for ∆T = 5.5 °C (l/h)
2600
Burner capacity (kW)
25.1
Fuel consumption (m³/h)
2.65
RI PT
Value
SC
Parameter
Table 2. Correction factor for the cooling capacity (adapted from [18]).
–10.0 1.14 1.11 1.06 0.99 0.88 0.72 0.52
M AN U
0, 5, 10, 15 20 25 30 35 40 45
Brine temperature (°C) –7.0 1.15 1.13 1.09 1.03 0.94 0.81 0.63
–5.0 1.15 1.14 1.12 1.07 1.00 0.89 0.75
–2.0 1.16 1.16 1.14 1.10 1.04 0.95 0.81
TE D
Ambient temperature (°C)
0.0 1.17 1.17 1.16 1.14 1.08 1.00 0.87
Table 3. Thermodynamic properties of each material stream. T Stream 24.8 30.5 24.4 305.7 277.4 200.7 160.9 160.9 –4.7 –1.4
AC C
0* 1 2 3 4 5 6 7 8 9
m ˙
e
E˙
(kPa)
(kg/s)
(kJ/kg)
(kW)
– 0.2588 0.0024 0.2612 0.2612 0.2612 0.0092 0.0092 0.7702 0.7702
– 0 49152 86.7 72.3 38.5 753.2 102.2 5.7 4.5
– 0 119.5 22.7 18.9 10.1 6.9 0.9 4.4 3.4
EP
(°C)
P
101.3 101.3 195.1 101.3 101.3 101.3 633.7 633.7 301.3 301.3
* Environment conditions.
24
ACCEPTED MANUSCRIPT
Stream Natural gas Power Heat (steam @ 5.3 bar) Cooling (brine @ –5 °C)
energy
exergy
(kW) 114.4 26.0 19.0 9.2
(kW) 119.5 26.0 6.9 0.9
RI PT
Table 4. Energy and exergy of each energy stream.
Table 5. Performance parameters of each plant subsystem. ε
COP
(–) 0.227 – 0.920 – 0.393
(–) 0.218 – 0.674 0.158 0.275
(–) – – – 0.440 –
E˙ d
yd
SC
MT EP HRSG AC Plant
η
(kW) 70.9 3.8 2.9 5.9 83.5
(–) 0.593 0.032 0.024 0.042 0.699
M AN U
Subsystem
Table A1. Cost flow rate and specific exergy cost of each energy stream (fuel and products). Fuel/ Product
Natural gas Power Cooling (brine @ –4.7°C)
F P P
C˙
c
(USD/h) 7.01 10.8 1.91
(USD/kWh) 0.059 0.416 2.026
TE D
Stream
E˙ d
Subsystem
EP
Table A2. Exergoeconomic parameters of each plant subsystem.
(–)
AC C
(kW)
yd
MT EP HRSG AC
70.9 3.8 2.9 5.9
0.593 0.032 0.024 0.042
cf
cp
Z˙
(USD/kWh) (USD/kWh) (USD/h) 0.06 0.06 0.07 0.15
0.24 0.07 0.15 1.65
4.77 0 0.26 0.67
C˙ d
r
f
(USD/h)
(–)
(–)
4.16 0.22 0.20 0.74
3.13 0.20 1.10 10.2
0.534 0 0.562 0.475
25
ACCEPTED MANUSCRIPT
Figures
RI PT
Figure 1. Scheme of the small scale cogeneration plant.
Figure 2. Chiller generator: (a) original design; (b) modified for cogeneration.
SC
Figure 3. Variation of the steam pressure with power output.
M AN U
Figure 4. Variation of the cooling capacity with steam pressure.
Figure 5. Variation of the cooling capacity with chiller outlet temperature.
TE D
Figure 6. Variation of the coefficient of performance with steam pressure.
Figure 7. Variation of the coefficient of performance with chiller outlet temperature.
EP
Figure 8. Variation of the chiller exergetic efficiency with steam pressure.
AC C
Figure 9. Variation of the chiller exergetic efficiency with cooling temperature.
26
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT