Applied Energy 147 (2015) 40–48

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier .com/ locate/apenergy

Enhancing micro gas turbine performance in hot climates through inletair cooling vapour compression technique

http://dx.doi.org/10.1016/j.apenergy.2015.02.0760306-2619/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Tel.: +39 0471 017816; fax: +39 0471 017009.E-mail addresses: g.comodi@univpm.it (G. Comodi), massimiliano.renzi@unibz.it

(M. Renzi), f.caresana@univpm.it (F. Caresana), l.pelagalli@univpm.it (L. Pelagalli).

G. Comodi a, M. Renzi b,⇑, F. Caresana a, L. Pelagalli a

a Dipartimento di Ingegneria Industriale e Scienze Matematiche, Università Politecnica delle Marche, via Brecce Bianche, 60131 Ancona, Italyb Libera Università di Bolzano, Facoltà di Scienze e Tecnologie, piazza Università, 5, 39100 Bolzano, Italy

h i g h l i g h t s

� A test bench has been designed to testdirect expansion IAC technique to aMGT.� The COP of the chiller ranged between

2.2 and 2.5.� Electric power gain depends on

ambient conditions and it reached upto 8.5%.� Electric efficiency gain depends on

ambient conditions and it reached upto 1.6%.� Performance gains are higher in drier

climates and with more performingchillers.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 26 June 2014Received in revised form 1 February 2015Accepted 20 February 2015

Keywords:Micro turbinesInlet air coolingDistributed generationElectrical efficiencyHot climatesDirect expansion

a b s t r a c t

Microturbines (MGTs) are power generation devices showing very interesting performance in terms oflow environmental impact, high-grade waste heat and very low maintenance cost. One of the main issuesthat affect the output of MGTs is their strong sensibility to inlet air temperature. Both in literature and inpractical applications, several solutions have been applied to control the inlet air conditions and reducethe sensibility of this kind of machines to ambient conditions. One of the most interesting technology isthe refrigerating vapour compression technique. This solution has already been used for medium/largeGTs, but there are very limited inlet air cooling applications on MGTs and few experimental data aredocumented. This paper describes a test bench that has been designed to apply the direct vapour expan-sion technique to a 100 kWe MGT and reports the power and efficiency augmentation of the machinewhen operating in hot summer days.

The chiller was designed to treat the MGT’s air flow rate under specific working conditions and cool theinlet air temperature down to 15 �C. Thanks to the reduction of the inlet air temperature, the MGTshowed a benefit in terms of electric power gain up to 8% with respect to the nominal power outputin ISO conditions while the electric efficiency increased by 1.5%. Results indicate that an almost lineartrend can be obtained both in the electric power increase and in the electric efficiency increase as afunction of the inlet air temperature when the chiller operates under nominal working conditions.

G. Comodi et al. / Applied Energy 147 (2015) 40–48 41

When the IAC device operates at a higher temperature or a higher humidity than the design one, the gainis limited; in some working conditions with high relative humidity, most of the beneficial effect can evenbe lost.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Among the energy production devices that can be adopted inthe distributed generation (DG) market, Microturbines (MGTs)could play a significant role [1–4] since their electric output variesfrom 25 kW to 500 kW, which is a particularly interesting range forcogeneration applications in the service sector, households andsmall industry. Even if the electric efficiency of MGTs is generallysmaller than that of internal combustion engines [5,6] they canbe profitably applied in DG, thanks to their high power density,low pollutants’ concentration, low operation and maintenance(O&M) costs and multi fuel capability [7]. A significant limitationon the application of MGTs, especially in hot climates, is the strongdependence of their performance, namely the electric output andthe electric efficiency, on the ambient conditions, being the hotambient temperature the most affecting parameter. This character-istic, that is also recorded for larger GT, does not allow to exploitthe full potential of these machines which are normally rated inISO conditions (15 �C, 101.3 kPa, 60% RH) [8].

Literature is rich of documents reporting the influence of atmo-spheric conditions on industrial scale GTs’ performance. Severalworks evaluated the performance loss of these machines in hot cli-mates: depending on the size and the characteristics of the GT, theelectric power output can decrease by 0.5–0.9%/�C [9–11]; evengreater losses are recorded for GTs of smaller capacities as reportedin a paper by Mohanty [12]. Amell and Cadavid [13] also investi-gated this issue and they attributed this behavior of smaller GTsnot only to the air density reduction with higher ambient tempera-ture but also to a volumetric flow reduction.

As for MGTs, the effect of the inlet air temperature on theirperformance has been investigated in a limited number of works[14–18]. This specific problem was already approached by theauthors of this paper: we designed a specific test bench to evaluatethe cogeneration performance of MGTs [19] and we adoptedthe artificial neural networks (ANNs) methodology [20] as well asanalytical models [21] to quantify the effect of ambient conditionson the output of the machine. Air pressure and relative humidity donot affect significantly the performance of the machine while airtemperature strongly affects both the electric output and the elec-tric efficiency. In numerical terms, a reduction of about 1.22%/�C forthe electric power and a reduction of about 0.51%/�C for the electricefficiency was assessed for Turbec100 kWe MGT if compared to theISO ratings.

Since the performance of GTs is so sensible to ambient tempera-ture variations, the so-called Inlet Air Cooling (IAC) techniqueshave been studied and applied to reduce its impact. There are sev-eral works in literature that presented a series of solutions and thebeneficial results that can be obtained by reducing the compressorinlet air temperature.

A work by Al-Ibrahim et al. [22] describes the most used IACtechniques that can be applied to enhance the performance ofGTs: (i) wetted media evaporative cooling; (ii) high-pressurefogging; (iii) absorption chiller cooling using the GT’s exhaustgas; (iv) and refrigerative vapour compression cooling.

In some cases, it is also possible to have a combination of theabovementioned technologies and obtain hybrid solutions in orderto use the most performing one depending on the environmentalconditions. A study by Al-Ansary et al. [23] showed that, combining

vapour compression cooling and fogging technologies, it is possibleto meet the requirements of both dry and humid climates and opti-mize the effectiveness of the IAC technique. Of course a drawbackof this solution is that the initial cost and the complexity of theplant are increased.

Among all the possible IAC technologies, the high-pressure fog-ging system shows a good compromise in terms or effectiveness,pay-back period and application simplicity [24]. It is particularlysuitable for hot and dry climates where it is possible to exploitmaximally the advantage of the adiabatic saturation. On the otherhand, it is not possible to control the temperature of the air down-stream the fogging nozzles as it is limited by the wet bulb tem-perature of the ambient air. The characteristics of this technologyhave been analyzed in several works. Sanaye [25] developed ananalytical approach to evaluate the compressor map working pointwhen high-pressure fogging is applied to GTs and combined cycleplants: a significant enhancement in the net power output wasreported as well as a general trend of the compressor operatingpoint towards the surge line. Besides the inlet fogging upstreamthe compressor, Roumeliotis [26] also studied the water/steaminjection in the combustor and applied to several commercialGTs showing results on both performance augmentation andengine operability. As regards the Brayton regenerated cycles,Kim [27] reported on the chance of adopting the fogging techniqueto enhance the performance of low-compression ratio GTs.

As regards the absorption cooling technique, the chance ofadopting an absorption chiller fed by the GT’s exhaust to treatthe inlet air of the gas compressor was investigated by Najjar[28]. Khaliq [29] conducted an energetic and exergy analysis ofan absorption inlet cooling cogeneration plant with evaporativeafter cooling showing significant advantages with respect to theoriginal basic cycle. Popli et al. [30] compared the positive effectof evaporative cooling and absorption chiller to a GT installed inan oil and gas installation in Abu Dhabi. IAC techniques are alsoapplied to more complex combined cycles where the exhaust ofthe GT are used to feed a bottoming steam cycle: Yang [31] devel-oped an analytical method to evaluate the influence of fogging andabsorption cooling techniques on the performance of a combinedcycle plant; he also suggested a range of ambient air temperatureand humidity where the IAC technologies can be favorably applied.

With regard to vapour compression techniques, a specific workwas carried out by Chacartegui [32] that evaluated the energeticand economic advantage of applying direct expansion cooling toseveral commercial cogeneration GTs. Mohapatra [33] comparedthe positive effect of vapour compression and vapour absorptionchillers applied to a combined cycle plant, also evaluating the effectof ambient humidity on the performance of the IAC techniques.

Besides the theoretical evaluations of the benefits of the applica-tion of the above mentioned solutions, IAC techniques are increas-ingly applied in many installations, especially in hot climates. Theirapplication to commercial GTs has been investigated by Kitchenet al. [34] who also calculated the achievable capacity increase; adetailed discussion of the available cooling techniques and themain advantages and drawbacks of each of them were discussedby Giourof [35], De Lucia et al. [36], ASHRAE [37], and Anderpont[38]; finally, a design guide was proposed by Stewart [39].

Another interesting solution to enhance MGT output is hotwater or steam injection in the combustion chamber, even though

42 G. Comodi et al. / Applied Energy 147 (2015) 40–48

this technique does not entail the treatment of the inlet air. Leeet al. [40] studied injection of hot water or steam in a 30 kWeMGT for CHP applications. Steam is produced by recovering theheat content of the exhaust and simulations of the machine perfor-mance are evaluated in the case of injection before the recuperatoror before the combustion chamber: in both cases the authorsreported significant performance improvements.

Summarizing the performance gain achieved by the above men-tioned IAC techniques, the electric power augmentation achievedby using the evaporative cooling technology is in the range offew percentage points (maximum 5%); as regards the high pressurefogging the power output increase ranges between 5% and 10%depending on the inlet ambient humidity, and thus on the capacityof reducing the inlet air temperature. As regards absorption coolingand vapour compression cooling the electric power gain can beover 20% when the ambient conditions are particularly favorableas it is possible to reduce sensibly the compressor inlet air tem-perature even below the wet bulb temperature (even down to afew degrees over 0 �C). Anyhow, also in this case ambient humidityplays a significant role as in humid climates most of the thermalpower from the chiller is used to abate the latent heat with onlya minor benefit in terms of GT electric output.

While IAC techniques have been widely applied in large sizedGT, there is a lack of experience in the application of these tech-niques to MGTs. In a previous paper [41] the authors of this workhave already described the application of the fogging technique toa Turbec T100 HP MGT, demonstrating very good performanceimprovements (over 10%), even though the power output enhance-ment is strongly dependent on the ambient conditions, namely airhumidity.

In this work, the direct expansion mechanical vapour compres-sion technology is adopted to treat the process air of the same typeof MGT in order to reduce the sensitivity of the machine’s perfor-mance to the ambient conditions. This technology was chosensince it shows better energy recovery performance with respectto indirect expansion systems and because it allows to adjust thetemperature at the inlet of the MGT with a simple retrofitted con-trol, independently of ambient conditions. While the evaporationtechniques are restricted by the wet bulb temperature, vapourcompression machines can also reach lower temperatures withthe only limitation of the maximum chiller capacity.

The aims of the work are to: (i) describe the test benchdeveloped to apply the direct expansion vapour compression IACtechnology to a 100 kWe MGT; (ii) report the experimental resultson the performance of the machine when operating in hot ambienttemperature with the IAC; (iii) describe the advantages of thistechnology in terms of electric power and electric efficiency gainand thus fulfill the lack of relevant scientific literature in the appli-cation of the direct expansion vapour compression technology toMGTs.

The paper is organized as follows: in Section 2 the MGT underanalysis is presented; Section 3 reports the description of theexperimental setup and the design of the IAC system; Section 4illustrates the results of the test campaign; finally Section 5 reportsthe concluding remarks.

2. Description of the MGT

The machine that is used in this work to study the effect of theIAC vapour compression technique is a TurbecT100 HP microtur-bine. The thermodynamic cycle that is operated by the machineis a regenerative Bryton cycle having a compression ratio of about4.5. In cogeneration configuration, downstream the regenerator,the machine is equipped with a heat exchanger that recovers thethermal power of the exhausts. The turbine and the compressor

are radial machines mounted on a single shaft with the electricgenerator. Since the MGT operates at very high rotational speedfor the small dimension of the radial machines, electricity isproduced at a voltage and at a frequency that are different fromthe standards of the grid. An electronic conversion system is thenable to convert the generated electricity to the correct valuesrequired by the electric grid.

As already mentioned, a test bed has been designed anddescribed by the authors of the present paper with the aim toevaluate the performance of the MGT in several working condi-tions [19]. The focus of the work was to evaluate the thermaland the electric output of the machine and to calculate thecorresponding overall efficiency. This test bench has been used alsoto study the effect of the IAC technique on the performance of theMGT as it will be described in the next paragraph.

3. The vapour compression chiller and the test bench

The vapour compression machine, a traditional industrialchiller that uses R507 as working fluid, was designed to reach anair temperature of about 15�C at the compressor’s inlet, whichrepresents the nominal working condition of the MGT. In orderto reduce the electric power consumption of the system inverterswere installed on the compressor and the condenser cooling fan.The first permits to regulate the cooling capacity and thus the airtemperature at the MGT inlet by varying the refrigerant mass flowrate, the second is used to regulate the compressor discharge pres-sure by varying the fan speed. The expansion valve is electronicallydriven.

3.1. Separation of the air flow

In the T100 microturbine a single duct supply the total inlet airflow which is about 1.6 kg/s. Once entered in the MGT cabinet, halfof the air flow is sucked by the compressor and it is used as theworking fluid of the gas turbine while the remaining part is usedto cool down the auxiliary systems. As a consequence, in order toapply the IAC technique to this MGT, it is necessary to split thetwo air flows so that only the air that actually enters the compressoris cooled down. To achieve this goal in the T100 MGT, a physicalseparation of the two air flows is realized by inserting a horizontalpartition wall that separates the upper area of the cabin, that facesthe inlet cone of the compressor, from the lower one where theauxiliaries’ cooling devices are located. In order to grant a sufficientcooling for the auxiliaries, new openings were realized in the lowerpart of the casing to facilitate the heat dissipation. Thanks to thisminimal modifications of the system, the two air streams enterthe MGT unit via two separate channels, thus allowing to operatethe IAC only in the duct that brings the working air to thecompressor.

3.2. Design of the vapour compression system installed in the testbench

The design of the vapour compression chiller was accomplishedconsidering the refrigeration capacity needed to take the nominalMGT working air flow rate, _mair = 0.8kg/s, from a reference ambientcondition of 30 �C and 60% R.H., to the set-point condition of 15 �C(nominal working temperature of the MGT) and 85% R.H. In theseconditions the air specific enthalpy changes fromhamb air ¼ 71:8kJ=kg to hcool air ¼ 31:9kJ=kg.

The correspondent required cooling capacity of the evaporatoris then calculated as:

Pc ¼ _mairðhamb air � hcool airÞ ð1Þ

yielding about 27 kW.

Table 1Main chiller components and their characteristics.

Component Manufacturer and model Characteristics

Compressor Bitzer 4DC-7.2Y power: 12 kW – with inverterAir condenser Alfa Laval ACS-502-C with 2.2 kWe inverter controlled cooling fansEvaporator KFL P35E working pressure: 26 barLiquid receiver FRIGOMEC 11 Volume: 11 lDry filter Carly DCY 165Thermal expansion valve Danfoss EVR10-32F-1214 Electronically driven

(a) (b)

Fig. 1. (a) Modified intake duct scheme and (b) direct expansion evaporator.

Table 2Measuring instruments used in the test bench.

Instrument Typology Accuracy

Air temperatureand humidity

SIEMENS QFM3171 Temperature: ±0.8 KHumidity: 2% R.H.

Pressure drop HUBA CONTROL 692 0.40%Fuel flow rate FLUENT ELITE Micro Motion 0.35%

G. Comodi et al. / Applied Energy 147 (2015) 40–48 43

The design conditions were chosen in order to make the MGTwork in nominal conditions for most of the days of a typicalmeteorological year in the location where the machine is installed.In fact, over-sizing the chiller with respect to the abovementionedconditions would be unprofitable and useful only in few days eachyear. In addition, the only drawback of a slight under-sizing of thechiller when the ambient temperature or ambient humidityexceeds the design parameters is that the MGT’s inlet temperatureeventually raises by a few degrees with respect to the chillerdesign conditions; this anyhow grants a significant recovery ofthe electric power loss and it might occur only few days a year.

As regards the practical realization of the chiller, the compres-sor, the condenser, the liquid receiver, the dryer filter and the ther-mal expansion valve are all enclosed in a casing; the characteristicsof the main components of the chiller are reported in Table 1. Thecompressor has a nominal maximum electric power of 12 kWe.The lamination valve is electronically driven and retrofitted bythe temperature of the refrigerated air in the MGT’s inlet duct.The cooling fluid is then sent to the evaporation unit that is locatedseparately from the rest of the system, inside the MGT’s air suctionchannel.

The condensed water on the evaporator heat exchanger isremoved by means of a specific system that will be described inthe next paragraph.

3.3. Modified intake duct for the evaporation heat exchanger and themeasurement instruments

In order to perform the experimental measurements, the intakeduct for the air elaborated by the compressor was modified anddesigned to embed the chiller’s evaporator system and the measur-ing apparatus. Fig. 1a shows the sketch of the modified intake ductwith the evaporation heat exchanger and the measuring appara-tus; Fig. 1b shows a picture of the chiller evaporator installedinside the inlet air duct.

The filter at the inlet of the fresh air duct is required to removethe coarse dust particles present in the intake air. The duct has a

square section (610 � 610 mm) in which the evaporator heatexchanger and the measuring probes are installed. In order toevaluate the condition of the air before and after the cooling treat-ment, temperature and relative humidity probes are installed asreported in Fig. 1a. In between the inlet section of the air ductand the inlet section of the MGT a differential pressure devicehas also been installed to evaluate the overall pressure lossesdue to the addition of the cooling system. The characteristics andthe accuracy of the installed probes are reported in Table 2.

The evaporator (see Fig. 1b) consists of a battery having copperpipes and aluminum fins and it is provided with a droplet waveseparator; the condensate is removed by gravity through a drainpipe.

All the other quantities required to evaluate the performance ofthe MGT, such as the electric power output, the fuel flow and theelectric efficiency of the machine are measured and acquired usingthe same procedure as the one reported in a previous paper of theauthors [19].

4. Results and discussion

The experimental data were collected during selected hot sum-mer days when the operation of the MGT is penalized by the highambient temperature. The level of relative humidity in the locationwhere the test bench has been realized during the test days rangedbetween 45% and 70% while ambient temperature ranged between23 �C and 34 �C. This means that some of the tests were performedwith ambient conditions that were more severe (higher tempera-ture and higher humidity) with respect to the design conditions

0

5

10

15

20

25

30

35

0

10

20

30

40

50

60

70

80

90

100

14:15 14:29 14:44 14:58 15:12

Evap

orat

or th

erm

al p

ower

[kW

]

Air

tem

pera

ture

[ºC

] / A

ir hu

mid

ity [%

]

Time

Air temperature before the HE

Air tempearture after the HE

Air humidity before the HE

Air humidity after the HE

Direct expansion HE thermal power

Fig. 2. Trend of the temperature and the relative humidity upstream and downstream the evaporator heat exchanger (HE) for a selected test day (initial relative humidity53%).

0

5

10

15

20

25

30

35

0

10

20

30

40

50

60

70

80

90

100

15:57 16:00 16:03 16:05 16:08

Evap

orat

or th

erm

al p

ower

[kW

]

Air

tem

pera

ture

[ºC

] / A

ir hu

mid

ity [%

]

Time

Air temperature before the HE

Air tempearture after the HE

Air humidity before the HE

Air humidity after the HE

Direct expansion HE thermal power

Fig. 3. Trend of the temperature and the relative humidity upstream and downstream the evaporator (initial relative humidity 45%).

44 G. Comodi et al. / Applied Energy 147 (2015) 40–48

of the chiller. In the following data are reported both for on-designand off-design working conditions.

During the tests, the pressure drop across the evaporator variedfrom 250 and 300 Pa, with a negligible effect on the machine per-formance [20].

Fig. 2 reports the trend of data measured by the air temperatureand air humidity probes in a selected test day. Data are reportedstarting from few minutes before activating the chiller andthroughout all the steady operation of both the chiller and theMGT. The graph clearly shows that when the IAC system operates,temperature drops down to 15 �C starting from 28 �C: the blue1

line refers to the air temperature before the evaporator heat exchan-ger (HE) and the red line is the temperature after it. Downstream theheat exchanger, relative humidity rises to 85–88% starting from avalue of about 52–54% (black and purple line respectively). The airtemperature reduction is effective when an approximate steady-state working condition is reached. The green line indicates the ther-mal power that is extracted by the evaporator in the inlet air duct,

1 For interpretation of color in Figs. 2–7, the reader is referred to the web version ofthis article.

both in terms of sensible and latent heat; the peak extracted thermalpower is about 20 kW.

A similar trend is also indicated in Fig. 3 that reports the sameacquired quantities in another selected test day. In this case ambi-ent humidity was lower, about 45%, and the ambient temperaturewas higher (about 33 �C). The evaporator extracts a maximum ofabout 27 kW of thermal power from the inlet air which reaches atemperature of 16.5 �C which is slightly higher than the targetone. As the ambient humidity has a significant weight on the valueof the specific enthalpy, it is possible to see that in a dry climate thethermal power required to cool down the stream of air is lowerthan the previous case where the humidity was higher.

In Fig. 4, besides the trend of the temperature upstream anddownstream the evaporator (red and blue lines respectively), thevalue of the power output of the MGT is depicted for the same testof Fig. 2. The graph reports the actual power produced by the MGTduring the tests with the IAC system operating (dark blue line) andthe trend of the electric power of the machine without the applica-tion IAC technique (purple line). This latter value was evaluatedusing the artificial neural network (ANN) methodology that hasalready been successfully applied by the authors of the presentpaper in the study of this MGT [20]. By means of this technique,

0

5

10

15

20

25

30

35

40

45

50

0

10

20

30

40

50

60

70

80

90

100

14:18 14:32 14:47 15:01

Air

tem

pera

ture

[ºC

]

Elec

tric

pow

er [k

W]

Time

MGT power with IAC

Electric power without IAC

Net electric power with IAC

Chiller electric consumption

Air tempearture after the HE

Air tempearture before the HE

Fig. 4. Inlet air temperature, MGT output and chiller power consumption with and without the IAC (initial humidity 53%).

10

15

20

25

30

35

40

45

50

0

10

20

30

40

50

60

70

80

90

100

15:57 15:59 16:01 16:04 16:06 16:08 16:10

Air

tem

pera

ture

[ºC

]

Elec

tric

pow

er [k

W]

Time

MGT power with IAC

Electric power without IAC

Net electric power with IAC

Chiller electric consumption

Air tempearture after the HE

Air tempearture before the HE

Fig. 5. Inlet air temperature, MGT output and chiller power consumption with and without the IAC (initial humidity 45%).

G. Comodi et al. / Applied Energy 147 (2015) 40–48 45

the maximum electric power output of the machine, as well as itselectric efficiency, can be calculated as a function of the inlet airtemperature, pressure and humidity and the machine load. In thefollowing sections of this paper, all the trends referring to themachine’s performance without the IAC system will be evaluatedusing this methodology and considering the air conditions col-lected by the probes installed upstream the evaporator. In addition,in the graph in Fig. 4 the orange line indicates the electric powerconsumption of the compression chiller, while the black line isobtained as the net electric power output of the system (MGTactual power output minus the chiller consumption). The electricpower gain obtained thanks to the application of the air treatmentis particularly significant and it reaches values over 14 kWe. Interms of net power output the gain is lower, due to the chiller con-sumption, and it can be evaluated in as much as 6 kWe. The trendof the MGT output clearly follows the trend of the air temperaturereduction in the inlet duct. When the compression chiller is acti-vated, the net power output has a decreasing threshold till thesteady state temperature condition is achieved in the inlet air.

Fig. 5 reports the same data and results corresponding to thetest day of Fig. 3 in which the value of the ambient temperaturewas higher; the electric power consumption of the compression

chiller results to be higher and the net electric power aug-mentation is lower. Also in this case, the net electric power outputincreases by 6 kWe with respect to the simple MGT without IACsince both the chiller has a higher consumption and the MGT hasa higher power recovery.

All the presented data referring to the test days were collectedtogether in the graph in Fig. 6: the gains in terms of electric effi-ciency and electric power output using the IAC system are pre-sented. During the tests ambient temperature varied from aminimum value of 22.5 �C to a maximum value of 33.5 �C. The bluedots show the MGT’s power output when the evaporation coolingchiller was active; the red dots report the electric output of themachine in absence of the IAC system. The first outcome of theresults is that, while in absence of a cooling device the machine’selectric output is strongly sensitive to ambient temperature, themachine performance is almost constant when the evaporationcooling is active; the only limitation is given by the capacity ofthe evaporator to bring the inlet air to the nominal set-point valueof 15 �C. Indeed, when ambient temperature and ambient humid-ity are higher than the design values (30 �C and 60% respectively),the chiller is no longer capable to cool down the air temperaturetill the design set-point and the MGT’s output shows a slight

20%

22%

24%

26%

28%

30%

32%

34%

0 102030405060708090

100110

21 24 27 30 33 36

Effic

ienc

y

Elec

tric

pow

er [k

W]

Ambient air temperature [ºC]

MGT electric power with IAC MGT electric power without IAC Chiller electric consumption MGT electric efficiency with IACElectrtic Efficiency without cooling

Fig. 6. Electric power, efficiency and chiller consumption as a function of the ambient temperature.

-2

0

2

4

6

8

10

-14-12-10-8 -6 -4 -2 0 2 4 6 8 1012 14

21 24 27 30 33 36

Effic

ienc

y ga

in [%

]

Elec

tric

pow

er g

ain

[kW

]

Ambient air temperature [ºC]

Net electric power gain

Net electric power gain off-design

Net electric efficiency gain

Net electric efficiency gain off-design

Fig. 7. Net electric power and efficiency gain of the system as a function of the ambient temperature and for two values of the chiller’s COP.

46 G. Comodi et al. / Applied Energy 147 (2015) 40–48

reduction. Ambient humidity has a minor effect on the electricoutput of the machine, as reported in [20] by the authors of thepresent paper.

In order to compare the performance of the system in presenceor in absence of the cooling system, in Fig. 6 the green dots reportthe electric consumption of the vapour compression chiller: ingeneral, it is possible to figure a linear increase of the electric con-sumption of the system with increasing ambient temperature; themean value of the COP ranged between 2.2 and 2.5 depending onthe working conditions (cooling fluid evaporation temperature)and the ambient temperature (condensing temperature), as it canbe noted by comparing Figs. 2 and 4, as well as Figs. 3 and 5.Nevertheless, it is worth noting that, in an evaporation coolingdevice, a significant part of the thermal power is required not onlyto cool down the treated air (sensible heat) but also to condensatethe air humidity (latent heat). Therefore, in the graph it is possibleto identify a cluster of data (highlighted with a red circle) that referto a test day in which the ambient humidity was particularly high,exceeding 60%. In this situation, even though the air temperaturedownstream the evaporator is almost close to the set-pointtemperature of 15 �C, the consumption of the chiller results to besignificantly higher with a direct impact on the system net outputin terms of electric power and electric efficiency. This aspect is

clearly visible in the data series reporting the MGT’s electricefficiency in Fig. 6. The purple dots indicate the MGT’s electricefficiency when the cooling device is active while the black dotsrefer to the net electric efficiency, calculated on the basis of thenet electric power output of the system (difference of the MGT’soutput and the chiller electric consumption) and the actualmachine’s fuel consumption. It is possible to conclude thatmachine’s efficiency is almost constant at 29% when the chilleroperates, mirroring the trend of the electric power output of theMGT (blue dots). Only a minor reduction can be recorded at highambient temperature when the evaporator is not able to supplythe whole required cooling effect. On the contrary, the net effi-ciency shows a much higher dispersion and it is also influencedby the electric consumption of the chiller. It is evident that, withhigh ambient humidity, the IAC system requires a higher electricpower which reflects on the net performance of the whole system.In particular, the value of the net electric efficiency ranges from amaximum value of about 27.3% at low ambient temperature to aminimum value of 24.2% at high ambient temperature and highambient humidity.

Finally, Fig. 7 reports both the electric power and the electricefficiency gains as a function of the ambient temperature for thewhole available set of experimental data. The grey and the light

G. Comodi et al. / Applied Energy 147 (2015) 40–48 47

blue data points refer to IAC results carried out with the chilleroperating under the design conditions described in Section 3.2;instead, the black and the dark blue dots refer to IAC results carriedout with the chiller operating under off-design conditions.

The net power increase (MGT power production minus thechiller power consumption, grey dots) ranges between 0 and8.5 kW depending on the ambient conditions, namely temperatureand humidity, while the electric efficiency gain ranges between�0.3% and 1.6% (light blue dots). In particular, the lower theambient humidity, the higher the sensible thermal power thatcan be extracted by the evaporator; therefore, for a given requiredtemperature reduction till the set point temperature of 15 �C, thechiller has a lower electric consumption and the IAC technique ismore performing. Conversely, when the ambient enthalpy is higherthan the design point (either given by higher temperature or ambi-ent humidity), the temperature at the MGT inlet is higher than theset point and the power increase is smaller, as well as the IAC tech-nique effectiveness. This effect is reported in Fig. 7 with the blackdots data for the electric power gain and the dark blue dots data forthe electric efficiency gain. These data points are highly dispersedsince the final electric performance of the system depends on thespecific ambient parameters. Therefore, in those conditions theelectric output enhancement is significantly reduced and the elec-tric efficiency gain can even be negative, down to �0.8%. Anotherinteresting remark can be deduced by the general trend of theelectric efficiency gain: as soon as the chiller works under designconditions and it is capable of cooling the air to the set point value,it is possible to figure an increasing trend of the electric efficiencygain. On the contrary, when the system works in off-design condi-tions, the chiller continuously operates at its nominal maximumelectric consumption but it is not able to cool down the air till15 �C anymore. Therefore, the higher the ambient temperature orthe ambient humidity, the higher the MGT’s inlet temperature;as a consequence, the net electric efficiency gain tends to decreasewith high ambient temperature.

On the basis of the reported data it is possible to sum up somecomments on the application of vapour compression cooling sys-tems coupled with a 100 kWe MGT: (i) the advantage in terms ofboth electric power gain and electric efficiency gain is very smallif the temperature difference achieved with the IAC device is lim-ited (for example when the air temperature is lower than 25 �Cand the set-point temperature is 15 �C); (ii) the level of the ambi-ent air humidity strongly influences the effectiveness of the IACsystem as a significant part of the thermal power exchanged inthe evaporator is used to reduce the latent heat and it does notresult in an increase of the MGT’s output; (iii) when the chilleroperates in off-design conditions, i.e. with higher temperatureand humidity with respect to the design ones, the temperaturedownstream the evaporator is higher than the nominal one, theelectric output gain tends to flatten and the electric efficiency gaindecreases.

5. Conclusions

The performance of MGTs is particularly sensitive to the ambi-ent conditions and hot inlet air temperature determines a signifi-cant loss in terms of electric power output, even higher thanlarge sized GTs as documented in several works. The vapour com-pression IAC technology can be used to reduce the MGT’s compres-sor inlet temperature and achieve material advantages in terms ofmachine performance in hot and dry climates. This solution,already proposed in large production plants, has never beenapplied to MGTs. The paper describes the design and the operationof a direct expansion IAC system that was applied to a 100 kWemicroturbine. The disposition and the dimensioning of the

evaporator of the chilling system were reported and the issuesarising from the necessity of separating the compressor workingfluid air and the auxiliaries’ cooling air were discussed. The set-point temperature of the air downstream the evaporator is set to15 �C which is the nominal working condition of the MGT.Experimental data were collected during hot summer days withambient temperature ranging between 23 and 34 �C and ambienthumidity ranging between 45% and 65%; in these conditions, thedirect expansion IAC system is particularly effective and it allowsto obtain remarkable advantages in terms of electric power outputand efficiency, especially when the ambient humidity is lower.

The performance of the proposed system is documented andthe advantage in terms of inlet air temperature reduction isreported. Specifically, thanks to the evaporator enclosed in theMGT’s air inlet channel, a maximum temperature reduction ofabout 16 �C was achieved when the ambient temperature was par-ticularly hot. The effect of the relative air humidity on the perfor-mance of the MGT with the IAC system is analyzed: when theambient humidity is high, a greater part of the evaporator coolingpower is required to condense the water content of the inlet air.During the tests it was possible to register a net electric power aug-mentation up to 8.5% referred to the machine’s ISO condition out-put; also the net electric efficiency gain showed a peak of 1.6%.Obviously, these figures can be further raised if a chiller withhigher COP is adopted. The reported test bench data indicate thatan almost linear trend can be obtained both in the electric powerincrease and in the electric efficiency increase as a function ofthe inlet air temperature when the chiller operates under nominalworking conditions. When the IAC device operates at higher tem-perature or humidity the gain is limited; in some working condi-tions with high relative humidity, most of the beneficial effectcan be lost. Finally, the flexibility of the described test bench willallow to deepen the study on the performance augmentation ofMGTs by regulating the operating parameters of the IAC systemdepending on the ambient conditions.

References

[1] Hwang Y. Potential energy benefits of integrated refrigeration system withmicroturbine and absorption chiller. Int J Refrig 2004;27:816–29.

[2] Wang W, Cai R, Zhang N. General characteristics of single shaft microturbineset at variable speed operation and its optimization. Appl Therm Eng2004;24:1851–63.

[3] Belmonte S, Núnez VN, Viramonte JG, Franco J. Potential renewable energyresources of the Lerma Valley, Salta, Argentina for its strategic territorialplanning. Renew Sustain Energy Rev 2009;13:1475–84.

[4] Ismail MS, Moghavvemi M, Mahlia TMI. Current utilization of microturbines asa part of a hybrid system in distributed generation technology. Renew SustainEnergy Rev 2013;21:142–52.

[5] Mousavi SMG. An autonomous hybrid energy system of wind/tidal/microturbine/battery storage. Electrical Power Energy Syst 2012;43:1144–54.

[6] Kalantar M, Mousavi SMG. Dynamic behavior of a stand-alone hybrid powergeneration system of wind turbine, microturbine, solar array and batterystorage. Appl Energy 2010;87(10):3051–64.

[7] Pepermans G, Driesen J, Haeseldonckx D, Belmans R, D’haeseleer W.Distributed generation: definition, benefits and issues. Energy Policy2005;33:787–98.

[8] ISO 2314. Gas-turbines – acceptance tests; 2009.[9] EI-Hadik. The impact of atmospheric conditions on gas turbine performance. J

Eng Gas Turbines Power 1990;112(4):590–6.[10] Kakaras E, Doukelis A, Prelipceanu A, Karellas S. Inlet air cooling methods for

gas turbine based power plants. J Eng Gas Turbines Power 2005;128(2):312–7.[11] Chaker M, Meher-Homji CB. Inlet fogging of gas turbine engines: climatic

analysis of gas turbine evaporative cooling potential of international locations.J Eng Gas Turbines Power 2006;128(4):815–25.

[12] Mohanty B, Paloso Jr G. Enhancing gas turbine performance by intake aircooling using an absorption chiller. Heat Recovery Syst CHP1995;15(1):41–50.

[13] Amell AA, Cadavid FJ. Influence of the relative humidity on the air coolingthermal load in gas turbine power plant. Appl Therm Eng 2002;22:1529–33.

[14] Basrawi F, Yamada T, Nakanishi K, Naing S. Effect of ambient temperature onthe performance of micro gas turbine with cogeneration system in cold region.Appl Therm Eng 2011;31:1058–67.

48 G. Comodi et al. / Applied Energy 147 (2015) 40–48

[15] Naing S, Yamada T, Nakanishi K. Applied performance research of acogeneration arrangement with proposed efficiency well-balance method. JPower Energy Syst 2007;1:199–210.

[16] Greenhouse Gas Technology Center. Environmental technology verificationreport: capstone 60 kW microturbine CHP system. Greenhouse Gas TechnologyCenter; 2003. <http://www.microturbine.com/_docs/EPA-C60testreport.pdf>.

[17] Midwest CHP Application Center and Avalon Consulting, Inc. Combined heat &power (CHP) resource guide, 2nd ed.;2005:11. <http://www.midwestcleanenergy.org/Archive/pdfs/Resource_Guide_10312005_Final_Rev5.pdf.

[18] U.S. Environmental Protection Agency Combined Heat and Power Partnership.Catalog of CHP Technologies. <http://www.epa.gov/chp/documents/catalog_chptech_full.pdf>.

[19] Caresana F, Comodi G, Pelagalli L, Renzi M, Vagni S. Use of a test-bed to studythe performance of micro gas turbines for cogeneration applications. ApplTherm Eng 2011;31:3552–8.

[20] Bartolini CM, Caresana F, Comodi G, Pelagalli L, Renzi M, Vagni S. Application ofartificial neural networks to micro gas turbines. Energy Convers Manage2011;52:781–8.

[21] Caresana F, Pelagalli L, Comodi G, Renzi M. Microturbogas cogenerationsystems for distributed generation: effects of ambient temperature on globalperformance and components’ behavior. Appl Energy 2014;124:17–27.

[22] Al-Ibrahim AM, Varnham A. A review of inlet air-cooling technologies forenhancing the performance of combustion turbines in Saudi Arabia. ApplTherm Eng 2010;30:1879–88.

[23] Al-Ansary HA, Orfi JA, Ali ME. Impact of the use of a hybrid turbine inlet aircooling system in arid climates. Energy Convers Manage 2013;75:214–23.

[24] Chacartegui R, Jiménez-Espadafor F, Sánchez D, Sánchez T. Analysis ofcombustion turbine inlet air cooling systems applied to an operatingcogeneration power plant. Energy Convers Manage 2008;49:2130–41.

[25] Sanaye S, Tahani M. Analysis of gas turbine operating parameters with inletfogging and wet compression processes. Appl Therm Eng 2013;30:234–44.

[26] Roumeliotis I, Mathioudakis K. Evaluation of water injection effect oncompressor and engine performance and operability. Appl Energy 2010;87:1207–16.

[27] Kim KH, Perez-Blanco H. Potential of regenerative gas-turbine systems withhigh fogging compression. Appl Energy 2007;84:16–28.

[28] Najjar YSH. Enhancement of performance of gas turbine engines by inlet aircooling and cogeneration system. Appl Therm Eng 1996;16:163–73.

[29] Khaliq A, Dincer I. Energetic and exergetic performance analyses of a combinedheat and power plant with absorption inlet cooling and evaporativeaftercooling. Energy 2011;36:2662–70.

[30] Popli S, Rodgers P, Eveloy V. Gas turbine efficiency enhancement using wasteheat powered absorption chillers in the oil and gas industry. Appl Therm Eng2013;50:918–31.

[31] Yang C, Yang Z, Cai R. Analytical method for evaluation of gas turbine inlet aircooling in combined cycle power plant. Appl Energy 2009;86:848–56.

[32] Chacartegui R, Jiménez-Espadafor F, Sánchez D, Sánchez T. Analysis ofcombustion turbine inlet air cooling systems applied to an operatingcogeneration power plant. Energy Convers Manage 2008;49:2130–41.

[33] Mohapatra Alok Ku, Sanjay. Thermodynamic assessment of impact of inlet aircooling techniques on gas turbine and combined cycle performance. Energy2014;68:191–203.

[34] Kitchen BJ, Ebeling JA. Qualifying combustion turbines for inlet air coolingcapacity enhancement. International gas turbine and aeroengine congress andexposition. Houston (Texas); 1995 June 5–8.

[35] Giourof A. Gas-turbine inlet-air cooling: you can almost pick your payback.Power 1995;139:56–8.

[36] De Lucia M, Lanfranchi C, Boggio V. Benefits of compressor inlet air coolingfor gas turbine cogeneration plants. J Eng Gas Turbines Power 1996;118:598–603.

[37] ASHRAE handbook HVAC systems and equipment; 2008. p. 17.1–5 [chapter17].

[38] Anderpont JS. Combustion turbine inlet air cooling (CTIAC): benefits andtechnology options in district energy applications. Trans Am Soc Heat RefrigAir Cond Eng 2001;107:892–902.

[39] Stewart WE. Design guide: combustion turbine inlet air cooling systems; 1999.[40] Lee JJ, Jeon MS, Kim TS. The influence of water and steam injection on the

performance of a recuperated cycle microturbine for combined heat andpower application. Appl Energy 2010;87:1307–16.

[41] Renzi M, Caresana F, Pelagalli L, Comodi G. Enhancing micro gas turbineperformance through fogging technique: experimental analysis. Appl Energy2015;135:165–73.