- Email: [email protected]
Effect of freeboard deflectors on the exergy in a fixed bed combustor
Applied Thermal Engineering 118 (2017) 62–72
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Effect of freeboard deflectors on the exergy in a fixed bed combustor Babak Rashidian ⇑, Yasir M. Al-Abdeli School of Engineering, Edith Cowan University, Joondalup, WA 6027, Australia
h i g h l i g h t s The influence of deflectors on the exergy analysis in a fixed bed biomass combustor is studied. Deflectors affect the axial mechanical exergy profiles inside the combustor. CO chemical exergy and thermomechanical exergy decrease when the k increases irrespective of deflectors. Deflectors don’t have a significant impact on the exhaust gasses availability.
a r t i c l e
i n f o
Article history: Received 22 August 2016 Revised 16 January 2017 Accepted 19 February 2017 Available online 22 February 2017 Keywords: Experimental data Fixed bed Biomass Combustion Freeboard deflector Exergy
a b s t r a c t Deflectors have been employed in industrial combustors and boilers with an expectation they reduce both radiation heat losses from the fuel bed and impact particle emissions. Despite much research into lab-scale biomass combustion, there have been no systematic studies to investigate the effects of deflectors on the axially resolved and flue gas availability in laboratory scale fixed bed biomass combustors. This study includes experiments conducted on a continuous feed pellet combustor, with a freeboard deflector located at different axial locations. The aim is to characterize the relative impact of freeboard deflectors on the mechanical exergy profiles and exhaust gas total exergy, over a range of stoichiometry (primary and secondary air flow rates). Results indicate that deflectors affect the mechanical exergy in the downstream, however their influence depends on their relative (axial) position (H). Furthermore, results reveal that for the tests with and without deflector, both CO chemical exergy and total exergy decrease in a similar manner when the air-fuel equivalence ratio (k) increases. It has been found that deflectors do not appear to affect the total and CO chemical exergy at the exhaust section of a labscale combustor, bearing in mind a ±3% variation in temperature, CO emissions as well as exergies is estimated based on the uncertainty analyses undertaken. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Biomass is a plentiful and well-utilised source of renewable energy which presents an opportunity to affect greenhouse gases emissions. Combustion of solid biomass fuels is a major technology in this regard which can help generate both heat and power [1,2]. In this context, grate-fired systems are widely used for biomass combustion in industrial plants [3], whereby moving grate systems are used in high thermal capacity industrial scale combustors [4]. Smaller scale combustors (<50 kW) incorporating fixed grates are also used [5]. However, in many combustion studies, laboratory scale combustors are used to investigate the effects of different process variables on combustion performance [6–8]. Because heat
⇑ Corresponding author at: School of Engineering, Edith Cowan University, 270 Joondalup Drive, WA 6027, Australia. E-mail address: [email protected] (B. Rashidian). http://dx.doi.org/10.1016/j.applthermaleng.2017.02.082 1359-4311/Ó 2017 Elsevier Ltd. All rights reserved.
and mass transfer in the vertical direction is considered as the dominating factor in a grate boiler, the results of the fixed bed laboratory scale biomass combustion models find some bearing to those in industrial moving bed [9], albeit with simplifying assumptions. Four parameters, none of which include exergy, are often used to quantify the performance of combustion in fixed beds [10]; the ignition front speed, burning rate, peak temperature and emissions. Ignition speed: Ignition speed which is also referred as flame front speed or reaction front velocity is based on the temperature change along the bed [11]. It shows the distance travelled by the reaction front per unit time [8]. Burning rate: The burning rate (ignition rate) is defined as the mass of fuel converted normalized to the combustor cross sectional and time (kg/m2 s) and expressed as:
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
_b¼ m
1 dM A dt
ð1Þ
In this regard, A is the cross section area of the fuel bed and dM/ dt is the change in fuel mass per change in time [8] and includes raw (moist) fuel conversion into both char or gas. To calculate the fuel mass conversion, combustor mass can be monitored by a scale or weight mechanism. Peak temperature: The localised temperature has many effects on combustion processes such as drying pyrolysis and gasification. The increase of the bed temperature results in an increase of fuel density, because chemical compounds with relatively low evaporation points, such as water and volatiles, are released to gas phase (vaporisation, devolatilization), leaving behind a solid phase which contains char and ash. Fig. 1 shows a typical relationship between the wood pellet density and temperature [8]. As such, the moisture content and solid fuel composition will vary continuously during combustion process as a function of bed temperature which also influences adiabatic combustion temperature. Emissions: Biomass is the most difficult to burn of the commonly used heating fuels. The amount of pollutants emitted to the atmosphere from different types of biomass combustion are highly dependent on the combustion technology implemented, the fuel properties and the combustion process conditions [12]. The amount of emissions released is one of the most important performance indicators in the biomass combustion. Many variables directly or indirectly influence emission levels
Density (kg/m3)
1600
1200
800 100
300
500
Temperature ( oC) Fig. 1. Variation of density with temperature [8].
such as bed temperature [8], fuel properties [13,14], air distribution rate and secondary air [15]. Table 1 shows some important characteristics of combustors used in the literature. However, in addition to the above four performance parameters, exergy conversion and efficiency are important to combustion in industrial applications of biomass, such as within power plants [16]. In this regard, the work producing potential of combustion products is a significant performance indicator for the combustion process [17] and is worthy of further study, as undertaken by this paper. The exergy analysis of combustion processes can play an important role in optimizing the design and operation of combustors. Exergy is a thermodynamic concept that describes the maximum work producing potential of a thermodynamic system when it is conducted into equilibrium reversibly with a reference environment [18,19]. Exergy analysis remains the subject of ongoing investigation in energy conversion units such as pulverized coal fired power plants [17], biomass gasifiers [20,21], internal combustion engines [22], industrial boilers [23,24], fluidized bed combustors [25], heat pump systems [26,27] and nuclear power plants [28]. An ideal process is reversible and for such process the exergy destruction is zero. Exergy loss occurs in practical processes due to thermodynamic irreversibilities even when there is no energy loss to the external environment [16]. Exergy consumption during a process is proportional to the entropy created due to irreversibilities [29]. Combustion processes are inherently irreversible which restrains the conversion of fuel energy into useful work. Chemical reactions and physical transport processes are the source of irreversibilities in combustion [30]. In typical atmospheric combustors about one-third of the fuel exergy is lost because of the inherent irreversibilities, which mostly come from heat transfer between products and reactants [16]. Fuel properties and the particular design and operational aspects of the combustion application are important factors that affect this. Woudstra et. al. [39] investigated the exergy efficiency for the combustion of different fuels, and found that exergy loss for the wood is higher compared to natural gas and coal. This may be because the process of biomass combustion is very complex and consists of many physical and chemical aspects including drying, devolatilisation, char burning and oxidization [6]. The main products of biomass combustion are particulate matter, gaseous emissions (CO, HC, NOx, SOx) as well as volatile organic compounds [40,41]. In commercial systems particle removal
Table 1 Wall material and insulations of reactors used in the literature.
a b c
Principal Author
Cross-Section
Diameter (mm)
Height (mm)
Liner Material
Insulation
Nicholls [31] Gort [32]
Circular Circular
510 300
1120 800
Refractory lining Stainless steel
Gort [32]
Circular
200
800
Stainless steel
Katunzi [8] Rogaume [33] Wiinikka [34]
Circular Circular Circular
56 200 200
500 2000 1700
Stainless steel –a Stainless steel
Van der Lans [35] Saastamoinen [36] Saastamoinen [36] Samuelsson [37] Weissinger [38] Yang [14] Ryu [13]
Circular Circular Square Square Circular Circular Circular
150 224 150 150 300 300 120 200 100
1370 300 900 700 300 1500 1500
Stainless steel – – – SiCb Inconelc Inconel
No insulation 2 mm ceramic fibre (inside), 100 mm glass wool (outside) 2 mm ceramic fibre (inside), 100 mm glass wool (outside) 40 mm glass wool 40 mm refractory layer, 40 mm rock fibre (outside) The initial 600 mm of the cylindrical walls from the bottom of the reactor have been insulated – – – Well insulated Loose external, casing of firebrick Thick layer insulating material in tight casing 80 mm insulation
No data available. Silicon carbide (SiC), also known as carborundum. A nickel-base alloy with chromium and iron.
63
64
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
devices such as electrostatic precipitators (ESPS), fabric filter systems, cyclones, afterburners, wet scrubbing equipment and deflectors are employed to reduce particle emissions [42–44]. Freeboard deflectors, which take the form of planer heat shields placed at some position above the freeboard, forms the main focus of this research. These devices are said to have two main functions in industrial. Firstly, they affect the draft force of the flue gases and increase residence time. Secondly, they affect temperature distributions in the freeboard zone [45], which can impact gaseous emissions [46]. Although numerous studies are available for the exergy analysis in combustion processes [21,25,29,47,48], no systematic work has been done into the effects of freeboard deflectors on the available exhaust gas power in fixed bed (continuous feed) laboratory scale biomass combustors. The overall aim of this study is to investigate the effects of freeboard deflectors on exergy loss in a laboratory scale fixed bed biomass combustor. For this purpose, experiments have been conducted on a continuous feeding system operated with, and without, a freeboard deflector. The deflector is placed at two different axial positions downstream of the fuel bed. This paper is a follow up on earlier works that look into the effects of freeboard deflectors in (non-reacting) packed beds [45] the fixed bed combustion of biomass [46], and the relevant experimental methodologies [49].
(a)
2. Methodology 2.1. Experimental methodology Experiments are conducted on a laboratory scale fixed bed continuous feeding combustor. Fig. 2a provides a diagram of the data acquisition and control installation. Fig. 2b also depicts the internal detail and four main zones: I- Plenum as well as II- Primary air, IIISecondary air, and IV- Exhaust sections. The overall height of the combustor is 1500 mm, with a 120 mm 120 mm square cross section and is made of high temperature resistant 310 stainless steel (6 mm thick). The exterior walls of the chamber are exposed to ambient conditions (i.e., uninsulated) so as to prevent overheating [15,50–52]. The operational (thermal) power of the combustor largely depends on the primary air flow rate (Qp), and is estimated to be around 15 kW. Both the primary and secondary air (Qs), which is induced further downstream, are supplied by two variable frequency drive centrifugal fans. Flow rates are measured with two mass air flow sensors. To maintain constant air flow rates the fan drives are coupled to a digital controller. Primary air is introduced from beneath through plenum and the grate. The secondary air section is placed at 300 mm above the grate and consists of 44 holes which are uniformly distributed along a single row and centred alongside the inner perimeter of this section.
(b)
Fig. 2. (a) Schematic of experiment setup; (b) CAD generated cross section of the combustor showing the internal future and deflector positions.
65
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Temperature measurements are made by fifteen thermocouples (Type N) coated by a Nicrobel protective metallic sheath. Each of these is 300 mm in length with outer diameter of 3 mm. The operating temperature range of thermocouples is from 270 °C to 1300 °C, at a resolution of 0.15 °C [53]. To resolve the temperature profiles, thermocouple probes (TC0-TC14) penetrate through access ports located at various axial positions along the combustor side wall. To ascertain radial temperature distributions, the thermocouples can be positioned radially by sliding them through each access point. Thermocouples TC0 to TC6 are used to resolve the downstream temperature profiles whilst thermocouples TC7 to TC14 determine upstream temperatures (where only Qp applies). A 16-Channel thermocouple module is used to feed data to an acquisition unit at a sampling rate of 0.2 Hz. In this study, standard cylindrical wood pellets with a 6 mm diameter and characteristic length of 20 mm are used [54]. Table 2 shows the proximate and elemental analysis of the fuel. The pellets are delivered from the hopper by a screw feeder system with a drive motor coupled to a manual on-off digital timer to control the fuel feeding rate. The hopper is suspended from a tensile load cell so as to monitor the mass loss of the fuel. The fuel pellets are fed upwards towards the inclined grate (25 degree gradient) which consists of 120 holes of 4 mm diameter, representing 40% open area (Fig. 2b). The primary air section consists of a detachable viewing port which is used for visual inspection of the combustion process and help a fixed fuel bed height. The exhaust section has an overall height of 1300 mm and located above the secondary air section. The square shaped deflector (95 mm 95 mm) is made of carbon steel and has a thickness of h = 5 mm. The deflectors axial position (H) can be regulated by an adjustable rod (with length of 400 mm) which is connected to a holding flange. The holder flange is installed 300 mm above the secondary air section and has an open area of 90%, whereby the adjusting rod passes through a bush hold by radial spokes which are laser cut into the holder flange. In this study, experiments have been performed for two different deflector positions. The deflector at H = 240 mm is positioned upstream of the secondary air, but between thermocouples TC7 and TC8. Other tests done with the deflector at H = 425 mm mean it is positioned downstream of the secondary air, between thermocouples TC5 and TC6. A continuous sampling gas analyser is used to measure CO, CO2, O2 and NO concentrations, with a suction probe located in the exhaust section at 110 mm above the last thermocouple (TC0). A LabView interface (Ver. 2014) is used to generate a time series of temperatures and emissions data. More details of the combustor and methodologies used in this study are described in previous publications [45,46,49].
not give information on the nature of internal losses and the quality of energy crossing the system boundary [55]. The maximum efficiency of a steady state process is limited by the second law of thermodynamics. The exergy (availability) of a system is defined as the maximum theoretical work that could be done by the system as it interacts with a specified reference environment and reaches thermal and chemical equilibrium [56]. The ensuing parts of this paper will present an exergy analysis for this fixed bed combustor. Complete high temperature oxidation of the biomass fuel occurs through combustion. During this process, the exergy of the biomass is converted mainly into internal energy of the hot flue gases. The chemical composition and the temperature of the gaseous reaction products depends on a number of factors such as the fuel composition, conditions of air supply (Qp, Qs) and heat loss to the surroundings. Fig. 3 shows a schematic diagram of the assumed control volume for the combustion chamber in a counter-current configuration [57]. Table 3 lists the reference environment properties which chosen in this study. In the pellet combustion process the ambient air is the environment. Variations in the ambient air intensive properties influence the system exergy even when the system’s states remain constant. This can be avoided by defining a reference environment with constant properties [29]. The compositions of the reference environment given in Table 3 are based on average concentrations for ambient air. The standard reference temperature and pressure of T0 = 298.17 K and P0 = 1.01325 105Pa have been defined for the reference environment.
Flue gas
CO, H2, CmHn, CO2, H2O, N2
Q Flame front
Fresh fuel
Biomass fuel
2.2. Exergy analysis
Air The first law of thermodynamics can help drive an energy balance for a system and determine heat and work transfer between the system and its environment. However, an energy balance does
Fig. 3. The assumed control volume governing thermal conversion in a countercurrent fixed bed configuration with continuous (fuel) feed.
Table 2 Nominal characteristics of the biomass fuel used [50]. Fuel type
Wood pellet
Formulaa
CH1.49O0.55
Ash analysis (wt.% of dry ash) Na2O MgO 6.22 10.36 a
Dry basis ash free.
Proximate analysis w.b. (%)
LHV (MJ kg1)
q (kg m3)
qP (kg m3)
Porosity
680
1240
0.45
Fe2O3 0.83
ZnO 0.24
Moist.
Vol.
Char
Ash
7.7
67.9
22.6
1.8
18.3
Al2O3 7.74
SiO2 16.47
P2O5 6.98
SO3 4.25
Cl 3.27
K2O 14.29
CaO 28.03
MnO 1.31
66
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Table 3 Reference environment properties [29].
Ambient air
Chemical composition
Mole fraction
N2 O2 H2O (g) CO2 Ar
0.7560 0.2034 0.0312 0.0003 0.0091
where ech p is the specific pellet chemical energy. Specific pellet chemical exergy has a constant value of 20,838,000 (J kg1) and varies proportionally to fuel consumption rate (kg s1) [29]. Carbon monoxide can be reacted with environmental oxygen to produce environmental carbon dioxide.
1 CO þ O2 ! CO2 2
ð7Þ
The chemical exergy of CO can be calculated by [19]:
Exergy consists of potential exergy (Eph), kinetic exergy (Ekn), physical exergy (Ep) and chemical exergy (Ec) [17,58]:
E ¼ Eph þ Ekn þ Ep þ Ec
ð2Þ
In a thermodynamic system such as an atmospheric pressure combustor where flow velocity and elevation changes have negligible effects [59], Ekn and Eph are not typically taken into account. For the remaining terms, Ep indicates the maximum theoretical useful work; therefore it is derived from the temperature and pressure differences between the system and its environment. The chemical non-equilibrium exergy Ec arises from chemical composition changes in the system [17]. Ekn, Eph and Ep can also be grouped together, naming the combination thermomechanical exergy, whilst the chemical non-equilibrium exergy is also named chemical exergy [29]. The thermomechanical exergy of a flow is defined by [58]:
etm ¼ ðh h0 Þ T 0 ðs s0 Þ
ð3Þ
where etm is the specific thermomechanical exergy (J/mol), h is the specific molar enthalpy, s is the specific molar entropy and T is the temperature. The subscript ‘‘0” indicates the reference state. The chemical exergy of a substance depends on its existence in the reference environment. In this research the molecule components of the pellet fuel, CO, N2, O2, H2O(g) and CO2, which participate in the elemental reaction of combustion, have been considered in the exergy analysis. Among these materials only the pellet and CO are not included in the considered environment (Table 3). Since the amount of the other emissions was not significant [46] only CO is taken into account. The general expression for the availability of flue gases is given by [19]:
where R is ideal gas constant, xk is the species mole fraction and x0 is the mole fraction of the element in reference environment. The first underlined term on the right side shows the contribution of temperature to the combustion gases availability, the second underlined term accounts for the chemical exergy (for the substances which are founded in the reference environment) and the third accounts for the pressure being different from P0. A biomass combustion process could be expressed as [29]:
b c b O2 ! aCO2 þ H2 O Ca Hb Oc þ a þ 4 2 2
ð5Þ
The chemical exergy of the pellet fuel is given by:
b c b h h ech ¼ h þ a þ ah p O CO h OðgasÞ p 2 2 4 2 2 2 ðT 0 ;Po Þ b c b sO2 asCO2 sh2 OðgasÞ T 0 sp þ a þ 4 2 2 ðT 0 ;P o Þ 2 3 b c aþ 0 4 2 ðxO2 Þ 5 þ RT 0 ln 4 2 a 0 ðxCO2 Þ ðx0H2 O Þ
1 1 h s ech ¼ h þ h T sh þ s CO O CO2 0 CO O CO2 CO 2 2 2 2 ðT 0 ;P o Þ ðT 0 ;Po Þ 2 3 1 0 2 ðxCO ÞðxO2 Þ 5 þ RT 0 ln 4 x0CO2
ð8Þ
where ech CO is the specific CO pellet chemical exergy. Thermomechanical exergy and chemical exergy can be grouped together, naming the total exergy and expressed as a molar basis as: ch e ¼ ef þ ech p þ eCO
ð9Þ
2.3. Test protocol Before igniting the fuel, biomass pellets are momentarily fed by the screw feeder to the grate and the bed height is set at 10–20 mm below the lowermost thermocouple TC14. After setting the bed height, the primary air is then adjusted and data acquisition started (t = 0 s) using the LabView software interface. With the viewing port removed, a heat gun is used to ignite the top layer of the solid fuel bed. As the flame spreads over top layer of the fuel bed, the viewing port window is reattached and the secondary air regulated. At this stage, the screw feeder driver motor is then once again activated and the mass flow of wood pellets into the combustor controlled by an on-off digital timer controller. The start-up period may take up to 1500 sec (Stage-I in Fig. 4). To self-extinguish the flame at the end of any test run, the screw feeder driver motor is stopped as well as both the primary and secondary supply of air. The shut-down period may take 600 sec (Stage-III in Fig. 4). Fig. 4 presents a time series for the evolution of downstream temperatures (TC0-TC7) and the rate of withdrawing pellets from the hopper for a single test. The steady state time period occurs within the start-up and shut-down (Stages II in Fig. 4). To characterize deflector effects on the exergy in this research, all the combustion parameters are extracted over the steady state time period for each test case. The method which is used to resolve steady state operating conditions can be found in [49]. A MATLAB (Ver. R2012b) script was developed to postprocess the data and calculate the flue gases availability. For exergy analysis of flue gases the average temperatures of TC0-TC2 which are very close to the emission measurement port (Fig. 2) have been used. The routine for calculating the exergy from the test cases presented, was checked against a worked example from literature [19]. Comparisons based on similar temperatures and species (CO) showed that the accuracy between the methods applied in this study and those in the other reference study where between ±2%. 2.4. Uncertainty analysis
ð6Þ
Every measurement is subject to some uncertainty [60]. Measurement errors can come from the measuring instrument such as thermocouple and gas analyser, from the operator, from the environment and from other sources. All experimental uncertainty is due to either random errors or systematic errors [60–62]:
67
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Fig. 4. Time series of downstream thermocouple data and residual fuel in the pellet hopper (Test 54).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
eTotal ¼ e2s þ e2r
ð10Þ
where eTotal is the total uncertainty, es is the systematic error and er is the random error. Systematic errors in experimental observations usually come from the measuring instrument limits (i.e. the accuracy of measurement device). Whilst random errors are statistical fluctuations in the measured data which caused by unknown and unpredictable changes in the experiment. These changes may occur in the measuring instruments or in the environmental conditions [61].
es ¼ er ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Xn
e
2 i¼1 s;i
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Xn
e
2 i¼1 r;i
ð11Þ ð12Þ
where n is the number of error source. The individual component of the random error in Eq. (12) is calculated as:
er;i
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi PN 2 rs i¼1 ð/i /Þ ¼ ¼ pffiffiffiffi NðN 1Þ N
N X ¼1 / / N i¼1 i
ð13Þ
ð14Þ
is the mean where rs is the standard deviation of a sample mean, / of a measured parameter and N is the number of samples in the repeated observations. 2.4.1. Systematic uncertainties Uncertainties in temperature and emission measurements are mainly associated with the accuracy of the thermocouples and gas analyser. Thermocouples used for measurements are Type N, (make: TC measurement, model: 2I-Nickel-Silicon-Magnesium) and are rated over a temperature range of 270 °C to 1300 °C. Within this range the temperature data acquisition system has an accuracy of 0.4% [53]. A continuous sampling gas analyser (make: Servomex, model: SERVOPRO 4900) is used to monitor fire species (O2, CO, CO2, NO) in the flue gas. Table 4 presents the accuracy of the gas analyser [63]. 2.4.2. Random uncertainty Random errors can be reliably estimated by repeating measurements. Measured data from tests 32, 33 and 34 have been chosen to estimate the random errors (Table 5). These test share same
operating conditions (Qp, Qs) and have similar combustion stoichiometry (0.43 < k < 0.51). More details can be found in Appendix A. 3. Results and discussion The primary air flow rates Qp result in an overall rich combustion stoichiometry (0.208 kprimary 0.719) in the upstream (TC0 to TC7, x = 0 mm to 265 mm). However, once the secondary air is introduced further down at x = 270 mm, the total air flow rates (primary + secondary) lead to lean combustion conditions (1.041 ktotal 3.599). This stoichiometry is more relevant when it comes to analysing the downstream flue gases. The relatively high ratio of secondary air (Qs/Qp) is introduced to prevent particulate (soot) deposition on Sections 2–4 of the combustor and their thermocouples. The experimental data which follows has banded into three ranges of stoichiometry (kprimary) and covers tests with deflectors at two axial distances (H = 240 and 425 mm) as well as baseline tests with no deflector (denoted ‘‘ND”). Table 5 presents the operating conditions of the selected tests. A sample error analysis for temperatures and emissions data has been applied to a limited number of test cases which feature largely similar conditions (Table 5, tests 32, 33 and 34) [10]. Whilst a much wider testing campaign featuring multiple repetitions of similar tests, is warranted so as to present a broader and more representative error analysis, Appendix A presents a summary of the uncertainty analysis applied to these three cases, whereby the likely ±3% variation on exergies has been estimated. Fig. 5 presents the normalized total exergy and CO chemical exergy for all the tests, with and without deflectors. Results are given for thermocouples placed at three radial positions (r = 5, 30 and 60 mm). The secondary horizontal axis shows the kTotal. Data presented is normalized by the peak exergy in each plot, for combustor operation either without a deflector (Fig. 5a) or with deflectors (Fig. 5b). The data clearly indicate the decline in both total exergy and CO exergy with increased k, whether the Table 4 Accuracy of the gas analyser. Gas
Accuracy
O2 CO CO2 NO
0.01% 1% 1% 1%
68
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Table 5 Operating condition of tests. Average fuel flow rate (gr/s)a
Deflector position, H (mm)
Thermocouple position, r (mm)
184 212 240 240 240 246 246 212 240 240 184
0.65 0.583 0.599 0.526 0.503 0.551 0.566 0.440 0.460 0.466 0.437
240 240 240 240 240 ND ND ND 425 ND ND
5 5 60 60 60 60 60 5 30 30 30
60 60 60 60 60 60 60 60 66 60 46 53 66 60 60 60
240 240 240 240 240 240 240 240 246 240 184 212 246 240 240 240
0.413 0.392 0.418 0.427 0.374 0.400 0.390 0.385 0.438 0.363 0.275 0.323 0.423 0.380 0.356 0.374
240 ND 425 425 240 425 ND ND ND 240 240 ND ND 425 240 ND
30 5 5 30 5 60 30 5 5 5 60 5 5 60 5 60
66 46 73 86 66 60 66 66 60 66 53 66 66 60 73 53
264 184 292 344 246 240 246 246 240 246 212 246 246 240 292 212
0.410 0.275 0.409 0.500 0.378 0.340 0.362 0.371 0.329 0.330 0.263 0.340 0.327 0.277 0.352 0.233
425 240 ND ND ND 425 240 240 425 240 240 425 ND ND ND ND
5 60 60 5 5 60 5 5 5 60 60 60 30 60 60 30
kprimary
Test
Q
k = 0.20–0.43
0.208 0.287 0.317 0.357 0.357 0.377 0.378 0.382 0.423 0.423 0.437
59 60 45 31 37 49 5 52 24 13 16
46 53 60 60 60 66 66 53 60 60 46
k = 0.43–0.51
0.439 0.439 0.441 0.447 0.461 0.475 0.475 0.475 0.478 0.489 0.491 0.492 0.494 0.500 0.506 0.509
35 43 30 40 33 25 41 17 54 32 63 22 53 29 34 48
k = 0.51–0.719
0.510 0.537 0.542 0.545 0.554 0.570 0.591 0.592 0.607 0.607 0.610 0.615 0.640 0.691 0.704 0.719
58 55 12 19 21 26 62 61 23 65 64 56 51 42 6 50
P
(l/min)
Q
s
(l/min)
ND: No deflector. a Steady state operation regime.
deflector is present or not. The lower exergy observed at leaner conditions may be attributed to the falling (average) temperature in downstream locations. It can be seen that in tests conducted both with the deflector (Fig. 5b), the data points for both exergies appear tighter (closer) together compared to tests without the deflectors. The significance of this however requires further research. The reasons for this behaviour could be due to the effect of deflectors on the near wall temperatures and CO concentrations [46]. In order to observe the effects of the deflector on the mechanical exergy (which is due to temperature differences over the axial locations in combustor) over kprimary = 0.439–0.509, Fig. 6 presents axial distributions of mechanical exergy for freeboard deflectors placed either upstream of the secondary air at H = 240 mm (Fig. 6b) or downstream of the secondary air at H = 425 mm (Fig. 6c). In this analysis, the flue gas is assumed as (all) air, since the emissions data is acquired only at the gas analyser sample port which is located at the exhaust section of the combustor. The effective temperature is defined as the average of radial temperatures at each axial position. It can be seen that for tests conducted with deflectors both upstream and downstream of the secondary air, the presence of a deflector appears to decrease the mechanical exergy downstream of the secondary air (TC0-TC5, Fig. 6b and c)
when compared with test cases without a deflector (Fig. 6a). However the contribution of the mechanical exergy is minimal towards the total exergy in comparison to the chemical exergy. Deflectors located downstream of the secondary air at H = 425 mm have more impact on upstream mechanical exergy (TC0-TC5, Fig. 6c). It is clear from these results that the magnitude of change in downstream mechanical exergy (TC0-TC0) is sensitive to deflector distance (H). However, the presence of a deflector whether upstream or downstream of the secondary air does not have significant impacts on the upstream mechanical exergy (TC7-TC13, Fig. 6). It is believed the effect on downstream mechanical exergy is due to the impacts of deflector on the temperature profiles in the downstream [46]. To further illustrate the global effects of deflectors on the exergy, Fig. 7 shows the total and CO chemical exergy of the flue gases for several test cases with the same primary air (Qp) and secondary air (Qs), and sharing a narrow range of stoichiometry (kprimary = 0.439–0.509), both with and without deflector. The exergy analysis is based on the emission concentrations at the exhaust section and average temperatures near the gas analyser sampling port (TC0-TC2). It is clear that, using deflectors does not have a significant impact on the total exergy and CO chemical exergy over the range of conditions tested.
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
69
(a) Tests without deflector
(b) Tests with deflectors Fig. 5. Total exergy and CO chemical exergy for tests with and without deflectors, with temperature data at various radial locations.
Fig. 6. Axial mechanical exergy distributions (a) test without deflector; (b) tests with deflectors placed at H = 240 mm; (c) tests with deflectors placed at H = 425 mm (0.43 kprimary 0.51).
Whilst using a deflector generally increases the level of the CO emissions [46], the magnitude of change in CO emissions is not sufficient to appreciably influence the chemical exergy (Fig. 7b). Whilst, the data shows that freebord deflectors affect the axially resolved mechanical exergy along the downstream length of the
combustor (Fig. 6), their effect on the flue gas exergy appears negligible (Fig. 6a). This may be due to the fact that for temperatures up to 208 °C the contribution associated with composition is dominant, whilst above 208 °C the contribution associated with temperature increases with temperature [19]. In this study, the
70
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Fig. 7. Total and CO chemical exergy for several conditions sharing same Qp and QS, both with and without deflector (0.43 kprimary 0.51).
temperature range of exhaust gas is 180 °C to 220 °C (Fig. 4). As such, whilst freeboard deflectors do appears to influence the axially resolved mechanical exergy, the overall variation of the mechanical exergy changes, both with/without deflectors, is insignificant. 4. Conclusions This study focuses on the effects of freeboard deflectors on the flue gases availability. Experiments were conducted on a continuous feed laboratory scale fixed bed combustor featuring both primary and secondary air, with a freeboard deflector placed at different axial locations. An exergetic analysis was applied to evaluate both the axially resolved mechanical exergy as well as the fume exergetic content of flue gases at the exit plane. A (sample) uncertainty analysis has also been applied. The following conclusions have been drawn as a result of the range of test conditions in this research: Results show that both CO chemical exergy and total exergy decrease when the air-fuel equivalence ratio (k) increases. The decline in CO chemical exergy and total exergy show similar trends for tests both with and without freeboard deflectors. The axially resolved mechanical exergy is affected by the use of freeboard deflectors. The inclusion of a freeboard deflector appears to decrease the mechanical exergy profiles in the downstream (TC0-TC5). This may be due to the heat transfer effects associated with freeboard deflectors but required further investigations. The mechanical exergy is sensitive to the axial position of the deflector (H). By raising the deflector position further above the packed bed, it more significantly affects the mechanical exergy in the downstream. Results reveal the deflectors do not have significant impact on the total exergy and CO chemical exergy of the flue gases at the exhaust section. This may be happen because the contribution associated with mechanical exergy at exhaust section (T < 220 °C) is insignificant. An uncertainty analysis has been applied to a sample number of text cases, whereby the likely ±3% variation on temperature, CO emissions as well as exergies has been estimated. Whilst the results presented herein have been derived over a specific (limited) range of stoichiometry and data set size, the observations made do indicate that more research is needed to ascertain the overall merits of using freeboard deflectors on a wider range of stoichiometric and fixed bed combustor geometries.
Acknowledgements Ms. Araceli Regueiro Pereira is gratefully acknowledged for assisting with some aspects of the data acquisition undertaken at the University de Vigo (Spain). The PhD research project of Mr. Babak Rashidian is made possible by an Edith Cowan University International Postgraduate Research Scholarship (ECU-IPRS).
Appendix A Table A-1 shows the average values of thermocouples and gas species data extracted over the steady state period for the selected tests. The total, systematic and random uncertainty values for the temperatures and emissions are derived using Eqs. (10), (11) and (12), respectively. The results are presented in Table A-2. Table A-3 presents the uncertainty values of the chemical exergy, CO exergy and total exergy.
Table A-1 Steady state values of temperatures and emissions for the selected tests. Test 32
Test 33
Test 34
Qp (l/min) Qs (l/min) Deflector Position, H (mm) Thermocouple position, r (mm)
60 240 240 5
60 240 240 5
60 240 240 5
Temperature (°C)
TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8 TC9 TC10 TC11 TC12 TC13 TC14
229.461 264.932 254.252 273.605 282.109 294.271 314.590 519.970 445.363 460.501 460.939 480.386 555.232 490.465 510.844
239.696 282.338 271.090 278.596 287.914 309.605 347.011 506.943 432.517 432.494 412.158 419.533 509.101 419.734 475.846
198.075 228.245 219.128 227.167 233.783 250.093 268.958 506.456 460.959 477.918 479.803 503.755 568.176 451.352 526.413
Emissions
NO (vpm) O2 (%) CO (%) CO2 (%)
52.848 14.909 1.391 8.696
54.247 14.443 1.461 8.718
50.878 15.048 1.566 8.611
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72 Table A-2 Systematic, random and total uncertainty of measured parameters. Parameter
Systematic uncertainty (±%)
Random uncertainty (±%)
Total uncertainty (±%)
Temperature (°C)
TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8 TC9 TC10 TC11 TC12 TC13 TC14
0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4
4.60 5.03 5.04 5.15 5.23 5.12 5.96 0.71 1.50 2.36 3.65 4.38 2.69 3.68 2.42
4.61 5.05 5.05 5.16 5.24 5.13 5.97 0.81 1.56 2.40 3.67 4.40 2.72 3.70 2.45
Emissions
NO O2 CO CO2
1 0.01 1 1
1.51 1.01 2.82 0.31
1.82 1.01 2.99 1.05
Table A-3 Uncertainty values of chemical exergy, CO exergy and total exergy. Parameter
Systematic uncertainty (±%)
Random uncertainty (±%)
Total uncertainty (±%)
Chemical exergy CO exergy Total exergy
1.00
1.89
2.14
0.4 1.78
1.36 1.58
1.69 2.38
References [1] T. Bridgwater, Biomass for energy, J. Sci. Food Agric. 86 (12) (2006) 1755–1768, http://dx.doi.org/10.1002/jsfa.2605. [2] S. Van Loo, J. Koppejan, The Handbook of Biomass Combustion and Co-Firing, Earthscan, UK, 2008, pp. 38–50. [3] C. Yin, L. Rosendahl, S.K. Kaer, Grate-firing of biomass for heat and power production, Prog. Energy Combust. Sci. 34 (2008) 725–754. [4] C. Yin, L. Rosendahl, S. Kaer, S. Clausen, S. Hvid, T. Hille, Mathematical modeling and experimental study of biomass combustion in a thermal 108 MW grate-fired boiler, Energy Fuels 22 (2) (2008) 1380–1390. [5] Y. Shynya, M. Yukihiko, The Asian biomass handbook, A guide for biomass production and utilization, The Japan Institute of Energy, Tokyo, Japan, 2006, p. 207–222. [6] H. Khodaei, Y.M. Al-Abdeli, F. Guzzomi, G.H. Yeoh, An overview of processes and considerations in the modelling of fixed-bed biomass combustion, Energy 88 (2015) 946–972. [7] C. Wurzenberger, S. Wallner, H. Raupenstrauch, Thermal conversion of biomass: comprehensive reactor and particle modeling, AIChE J. 48 (10) (2002) 2398–2411. [8] M. Katunzi, Biomass conversion in fixed bed experiments, Report number WVT 2006.18, Eindhoven University of Technology, Department of Mechanical Engineering: Netherlands, 2006, p. 1–55. [9] Y.B. Yang, C. Ryu, A. Khor, N.E. Yatesb, V.N. Sharifi, J. Swithenbank, Effect of fuel properties on biomass combustion. Part II. Modelling approach-identification of the controlling factors, Fuel 84 (16) (2005) 2116–2130. [10] B. Rashidian, Deflector effects in fixed bed (biomass) combustors and noncombusting packed beds (Ph.D. thesis), Edith Cowan University, School of Engineering, 2016. [11] C. Ryu, A. Phan, Y.B. Yang, V.N. Sharifi, J. Swithenbank, Ignition and burning rates of segregated waste combustion in packed beds, Waste Manage. 27 (6) (2006) 802–810. [12] T. Nussbaumer, C. Czasch, N. Klippel, J. Johansson, C. Tullin, Primary and secondary measures for the reduction of nitric oxide emissions from biomass combustion, in: Developments in Thermochemical Biomass Conversion, Springer, Netherlands, 1996, p. 1447–1461. [13] C. Ryu, Y.B. Yang, A. Khor, N. Yates, V.N. Sharifi, J. Swithenbank, Effect of fuel properties on biomass combustion: Part 1. Experiments-fuel type, equivalence ratio and particle size, Fuel 85 (2006) 1039–1046.
71
[14] Y.B. Yang, V.N. Sharifi, J. Swithenbank, Effect of air flowrate and fuel moisture on the burning behaviors of biomass and simulated municipal solid wastes in packed beds, Fuel 83 (11) (2004) 1553–1562. [15] H. Wiinikka, R. Gebart, The influence of air distribution rate on particle emissions in fixed bed combustion of biomass, Combust. Sci. Technol. 177 (9) (2005) 1747–1766, http://dx.doi.org/10.1080/00102200590959468. [16] S. Som, A. Datta, Thermodynamic irreversibilities and exergy balance in combustion processes, Prog. Energy Combust. Sci. 34 (3) (2008) 351–376. [17] J. Xiong, H. Zhao, C. Zheng, Exergy analysis of a 600 MWe oxy-combustion pulverized-coal-fired power plant, Energy Fuels 25 (8) (2011) 3854–3864. [18] A. Bejan, M.J. Moran, Thermal Design and Optimization, John Wiley & Sons, 1996, pp. 140–150. [19] M.J. Moran, Availability Analysis: A Guide to Efficient Energy Use, PrenticeHall, New Jersey, 1982, pp. 130–160. [20] Y. Wu, W. Yang, W. Blasiak, Energy and exergy analysis of high temperature agent gasification of biomass, Energies 7 (4) (2014) 2107–2122. [21] K.J. Ptasinski, Thermodynamic efficiency of biomass gasification and biofuels conversion, Biofuels, Bioprod. Biorefin. 2 (3) (2008) 239–253. [22] H. Caliskan, A. Hepbasli, Exergetic cost analysis and sustainability assessment of an internal combustion engine, Int. J. Exergy 8 (3) (2011) 310–324. [23] R. Saidur, J. Ahamed, H. Masjuki, Energy, exergy and economic analysis of industrial boilers, Energy Policy 38 (5) (2010) 2188–2197. [24] M. Hasanuzzaman, N. Rahim, R. Saidurm Analysis of energy, exergy and energy savings of a fire tube boiler, in: 2011 IEEE First Conference on Clean Energy and Technology (CET), IEEE, 2011, pp. 291–296. [25] C. Martín, M.A. Villamañán, C.R. Chamorro, J. Otero, A. Cabanillas, J.J. Segovia, Low-grade coal and biomass co-combustion on fluidized bed: exergy analysis, Energy 31 (2) (2006) 330–344. [26] H. Esen, M. Inalli, M. Esen, K. Pihtili, Energy and exergy analysis of a groundcoupled heat pump system with two horizontal ground heat exchangers, Build. Environ. 42 (10) (2007) 3606–3615. [27] M. Esen, T. Yuksel, Experimental evaluation of using various renewable energy sources for heating a greenhouse, Energy Buildings 65 (2013) 340–351. [28] M.A. Rosen, Energy-and exergy-based comparison of coal-fired and nuclear steam power plants, Exergy Int. J. 1 (3) (2001) 180–192. [29] E. Granada, D. Patiño, J. Collazo, J. Moran, J. Porteiro, Available exhaust gas power in different configurations in a pellet stove plant, Renewable Energy 34 (12) (2009) 2852–2859. [30] W. Dunbar, N. Lior, Sources of combustion irreversibility, Combust. Sci. Technol. 103 (1–6) (1994) 41–61. [31] P. Nicholls, Underfeed Combustion, Effect of Preheat, and Distribution of Ash in Fuel Beds, Bulletin of the United States Bureau of Mines, 1934, p. 378. [32] R. Gort, On the propagation of a reaction front in a packed bed: thermal conversion of municipal waste and biomass (Ph.D. thesis), Universiteit Twente, Enschede, 1995. [33] T. Rogaumea, M. Auzanneau, F. Jabouille, J.C. Goudeau, J.L. Torerob, The effects of different airflows on the formation of pollutants during waste incineration, Fuel 81 (2002) 2277–2288. [34] H. Wiinikka, R. Gebart, Critical parameters for particle emissions in small-scale fixed-bed combustion of wood pellets, Am. Chem. Soc. J. 18 (2004) 898–907. [35] R.P. Van der Lans, L.T. Pedersen, A. Jensen, P. Glarborg, D. Johansen, Modelling and experiments of straw combustion in a grate furnace, Biomass Bioenergy 19 (2000) 199–208. [36] J.J. Saastamoinen, M. Horttanainen, P. Sarkomaa, Ignition wave propagation and release of volatiles in beds of wood particles, Combust. Sci. Technol. 165 (1) (2001) 41–60. [37] J.I. Samuelsson, Conversion of Nitrogen in a Fixed Buring Biofuel Bed Thesis for the degree of licentiate of engineering, Chalmers University of Technology, 2006. [38] A. Weissinger, T. Fleckl, I. Obernberger, In situ FT-IR spectroscopic investigations of species from biomass fuels in a laboratory-scale combustor: the release of nitrogenous species, Combust. Flame 137 (4) (2004) 403–417. [39] N. Woudstra, T. van der Stelt, Exergy analysis of combustion systems, in: Proceedings of the International ECOS Conference, Copenhagen, Denmark, June, 2003. [40] N. Klippel, T. Nussbaumer, Health relevance of particles from wood combustion in comparison to diesel soot, in: 15th European Biomass Conference, Berlin, 2007. [41] B. Kovacevik, Alkali-containing aerosol particles release during biomass combustion and ambient air concentrations (Ph.D. thesis), University of Gothenburg, Department of Chemistry, Atmospheric Science, 2009. [42] K. Porle, N. Klippel, O. Riccius, I.I. Kauppinen, T. Lind, Full scale ESP performance after PC-boilers firing low sulfur coals, in: Proceedings of EPRI/ DOE International Conference of Managing Hazardous and Particulate Air Pollutants, Toronto, Canada, 1997. [43] L. Salaver, T. Lee, C.D. Cooper, C. Yu Wu, J. Donnelly, Air Pollution Emission Control Devices for Stationary Sources, Air & Waste Management Association (A&WMA), 2007. [44] R. Buczyn´ski, R. Weber, A. Szle˛k, Innovative design solutions for small-scale domestic boilers: Combustion improvements using a CFD-based mathematical model, J. Energy Inst. 88 (1) (2015) 53–63. [45] B. Rashidian, Y.M. Al-Abdeli, G.H. Yeoh, F.G. Guzzomi, Effect of freeboard deflectors on the temperature distribution in packed beds, Appl. Therm. Eng. 89 (2015) 134–143.
72
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
[46] B. Rashidian, Y.M. Al-Abdeli, D. Patiño, F.G. Guzzomi, G.H. Yeoh, Effect of freeboard deflectors in the fixed bed combustion of biomass, Appl. Therm. Eng. 103 (2016) 543–552. [47] A. Abuadala, I. Dincer, G. Naterer, Exergy analysis of hydrogen production from biomass gasification, Int. J. Hydrogen Energy 35 (10) (2010) 4981– 4990. [48] T. Nussbaumer, M. Oser, Evaluation of Biomass Combustion Based Energy Systems by Cumulative Energy Demand and Energy Yield Coefficient, International Energy Agency and Swiss Federal Office of Energy, Zurich, 2004, January. [49] B. Rashidian, Y.M. Al-Abdeli, G.H. Yeoh, D. Patiño, F. Guzzomi, Methodologies for processing fixed bed combustor data, Combust. Sci. Technol. 189 (1) (2017) 79–102. [50] J. Porteiro, E. Granada, J. Collazo, D. Patiño, J. Morán, Experimental analysis of ignition front of several biomass fuel in a fixed-bed combuster, Fuel 89 (1) (2010) 26–35. [51] L. Febrero, E. Granada, D. Patiño, P. Eguía, A. Regueiro, A comparative study of fouling and bottom ash from woody biomass combustion in a fixed-bed smallscale boiler and evaluation of the analytical techniques used, Sustainability 7 (5) (2015) 5819–5837. [52] L. Febrero, E. Granada, A. Regueiro, J.L. Míguez, Influence of combustion parameters on fouling composition after wood pellet burning in a lab-scale low-power boiler, Energies 8 (9) (2015) 9794–9816. [53] NI 9213 Operating instructions and specifications, N.I. Corporation, Editor, Austin, USA, 2009.
[54] J. Porteiro, D. Patiño, J. Moran, E. Granada, Study of a fixed-bed biomass combustor: Influential parameters on ignition front propagation using parametric analysis, Energy Fuels 24 (7) (2010) 3890–3897, http://dx.doi. org/10.1021/ef100422y. [55] R. Prommas, P. Rattanadecho, D. Cholaseuk, Energy and exergy analyses in drying process of porous media using hot air, Int. Commun. Heat Mass Transfer 37 (4) (2010) 372–378. [56] I. Dincer, Y.A. Cengel, Energy, entropy and exergy concepts and their roles in thermal engineering, Entropy 3 (3) (2001) 116–149. [57] H. Thunman, B. Leckner, Co-current and counter-current fixed bed combustion of biofuel-a comparison, Fuel 82 (2003) 275–283. [58] N. Shigaki, T. Akiyama, F. Tsukihashi, Exergy analysis of steel production processes, Mater. Trans. 43 (3) (2002) 379–384. [59] P. Ahmadi, I. Dincer, Exergoenvironmental analysis and optimization of a cogeneration plant system using Multimodal Genetic Algorithm (MGA), Energy 35 (12) (2010) 5161–5172. [60] Z.U. Ahmed, An experimental and numerical study of surface interactions in turbulent swirling jets (Ph.D. thesis), Edith Cowan University, School of Engineering, 2016. [61] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Thermal Fluid Sci. 1 (1) (1988) 3–17. [62] D.C. Montgomery, G.C. Runger, Applied Statistics and Probability for Engineers, fifth ed., John Wiley & Sons, USA, 2010, pp. 370–380. [63] ServoPro Configuration and operation 4000 Series, S.G. Limited, Editor, East Sussex, UK, 2011.
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Effect of freeboard deflectors on the exergy in a fixed bed combustor Babak Rashidian ⇑, Yasir M. Al-Abdeli School of Engineering, Edith Cowan University, Joondalup, WA 6027, Australia
h i g h l i g h t s The influence of deflectors on the exergy analysis in a fixed bed biomass combustor is studied. Deflectors affect the axial mechanical exergy profiles inside the combustor. CO chemical exergy and thermomechanical exergy decrease when the k increases irrespective of deflectors. Deflectors don’t have a significant impact on the exhaust gasses availability.
a r t i c l e
i n f o
Article history: Received 22 August 2016 Revised 16 January 2017 Accepted 19 February 2017 Available online 22 February 2017 Keywords: Experimental data Fixed bed Biomass Combustion Freeboard deflector Exergy
a b s t r a c t Deflectors have been employed in industrial combustors and boilers with an expectation they reduce both radiation heat losses from the fuel bed and impact particle emissions. Despite much research into lab-scale biomass combustion, there have been no systematic studies to investigate the effects of deflectors on the axially resolved and flue gas availability in laboratory scale fixed bed biomass combustors. This study includes experiments conducted on a continuous feed pellet combustor, with a freeboard deflector located at different axial locations. The aim is to characterize the relative impact of freeboard deflectors on the mechanical exergy profiles and exhaust gas total exergy, over a range of stoichiometry (primary and secondary air flow rates). Results indicate that deflectors affect the mechanical exergy in the downstream, however their influence depends on their relative (axial) position (H). Furthermore, results reveal that for the tests with and without deflector, both CO chemical exergy and total exergy decrease in a similar manner when the air-fuel equivalence ratio (k) increases. It has been found that deflectors do not appear to affect the total and CO chemical exergy at the exhaust section of a labscale combustor, bearing in mind a ±3% variation in temperature, CO emissions as well as exergies is estimated based on the uncertainty analyses undertaken. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Biomass is a plentiful and well-utilised source of renewable energy which presents an opportunity to affect greenhouse gases emissions. Combustion of solid biomass fuels is a major technology in this regard which can help generate both heat and power [1,2]. In this context, grate-fired systems are widely used for biomass combustion in industrial plants [3], whereby moving grate systems are used in high thermal capacity industrial scale combustors [4]. Smaller scale combustors (<50 kW) incorporating fixed grates are also used [5]. However, in many combustion studies, laboratory scale combustors are used to investigate the effects of different process variables on combustion performance [6–8]. Because heat
⇑ Corresponding author at: School of Engineering, Edith Cowan University, 270 Joondalup Drive, WA 6027, Australia. E-mail address: [email protected] (B. Rashidian). http://dx.doi.org/10.1016/j.applthermaleng.2017.02.082 1359-4311/Ó 2017 Elsevier Ltd. All rights reserved.
and mass transfer in the vertical direction is considered as the dominating factor in a grate boiler, the results of the fixed bed laboratory scale biomass combustion models find some bearing to those in industrial moving bed [9], albeit with simplifying assumptions. Four parameters, none of which include exergy, are often used to quantify the performance of combustion in fixed beds [10]; the ignition front speed, burning rate, peak temperature and emissions. Ignition speed: Ignition speed which is also referred as flame front speed or reaction front velocity is based on the temperature change along the bed [11]. It shows the distance travelled by the reaction front per unit time [8]. Burning rate: The burning rate (ignition rate) is defined as the mass of fuel converted normalized to the combustor cross sectional and time (kg/m2 s) and expressed as:
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
_b¼ m
1 dM A dt
ð1Þ
In this regard, A is the cross section area of the fuel bed and dM/ dt is the change in fuel mass per change in time [8] and includes raw (moist) fuel conversion into both char or gas. To calculate the fuel mass conversion, combustor mass can be monitored by a scale or weight mechanism. Peak temperature: The localised temperature has many effects on combustion processes such as drying pyrolysis and gasification. The increase of the bed temperature results in an increase of fuel density, because chemical compounds with relatively low evaporation points, such as water and volatiles, are released to gas phase (vaporisation, devolatilization), leaving behind a solid phase which contains char and ash. Fig. 1 shows a typical relationship between the wood pellet density and temperature [8]. As such, the moisture content and solid fuel composition will vary continuously during combustion process as a function of bed temperature which also influences adiabatic combustion temperature. Emissions: Biomass is the most difficult to burn of the commonly used heating fuels. The amount of pollutants emitted to the atmosphere from different types of biomass combustion are highly dependent on the combustion technology implemented, the fuel properties and the combustion process conditions [12]. The amount of emissions released is one of the most important performance indicators in the biomass combustion. Many variables directly or indirectly influence emission levels
Density (kg/m3)
1600
1200
800 100
300
500
Temperature ( oC) Fig. 1. Variation of density with temperature [8].
such as bed temperature [8], fuel properties [13,14], air distribution rate and secondary air [15]. Table 1 shows some important characteristics of combustors used in the literature. However, in addition to the above four performance parameters, exergy conversion and efficiency are important to combustion in industrial applications of biomass, such as within power plants [16]. In this regard, the work producing potential of combustion products is a significant performance indicator for the combustion process [17] and is worthy of further study, as undertaken by this paper. The exergy analysis of combustion processes can play an important role in optimizing the design and operation of combustors. Exergy is a thermodynamic concept that describes the maximum work producing potential of a thermodynamic system when it is conducted into equilibrium reversibly with a reference environment [18,19]. Exergy analysis remains the subject of ongoing investigation in energy conversion units such as pulverized coal fired power plants [17], biomass gasifiers [20,21], internal combustion engines [22], industrial boilers [23,24], fluidized bed combustors [25], heat pump systems [26,27] and nuclear power plants [28]. An ideal process is reversible and for such process the exergy destruction is zero. Exergy loss occurs in practical processes due to thermodynamic irreversibilities even when there is no energy loss to the external environment [16]. Exergy consumption during a process is proportional to the entropy created due to irreversibilities [29]. Combustion processes are inherently irreversible which restrains the conversion of fuel energy into useful work. Chemical reactions and physical transport processes are the source of irreversibilities in combustion [30]. In typical atmospheric combustors about one-third of the fuel exergy is lost because of the inherent irreversibilities, which mostly come from heat transfer between products and reactants [16]. Fuel properties and the particular design and operational aspects of the combustion application are important factors that affect this. Woudstra et. al. [39] investigated the exergy efficiency for the combustion of different fuels, and found that exergy loss for the wood is higher compared to natural gas and coal. This may be because the process of biomass combustion is very complex and consists of many physical and chemical aspects including drying, devolatilisation, char burning and oxidization [6]. The main products of biomass combustion are particulate matter, gaseous emissions (CO, HC, NOx, SOx) as well as volatile organic compounds [40,41]. In commercial systems particle removal
Table 1 Wall material and insulations of reactors used in the literature.
a b c
Principal Author
Cross-Section
Diameter (mm)
Height (mm)
Liner Material
Insulation
Nicholls [31] Gort [32]
Circular Circular
510 300
1120 800
Refractory lining Stainless steel
Gort [32]
Circular
200
800
Stainless steel
Katunzi [8] Rogaume [33] Wiinikka [34]
Circular Circular Circular
56 200 200
500 2000 1700
Stainless steel –a Stainless steel
Van der Lans [35] Saastamoinen [36] Saastamoinen [36] Samuelsson [37] Weissinger [38] Yang [14] Ryu [13]
Circular Circular Square Square Circular Circular Circular
150 224 150 150 300 300 120 200 100
1370 300 900 700 300 1500 1500
Stainless steel – – – SiCb Inconelc Inconel
No insulation 2 mm ceramic fibre (inside), 100 mm glass wool (outside) 2 mm ceramic fibre (inside), 100 mm glass wool (outside) 40 mm glass wool 40 mm refractory layer, 40 mm rock fibre (outside) The initial 600 mm of the cylindrical walls from the bottom of the reactor have been insulated – – – Well insulated Loose external, casing of firebrick Thick layer insulating material in tight casing 80 mm insulation
No data available. Silicon carbide (SiC), also known as carborundum. A nickel-base alloy with chromium and iron.
63
64
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
devices such as electrostatic precipitators (ESPS), fabric filter systems, cyclones, afterburners, wet scrubbing equipment and deflectors are employed to reduce particle emissions [42–44]. Freeboard deflectors, which take the form of planer heat shields placed at some position above the freeboard, forms the main focus of this research. These devices are said to have two main functions in industrial. Firstly, they affect the draft force of the flue gases and increase residence time. Secondly, they affect temperature distributions in the freeboard zone [45], which can impact gaseous emissions [46]. Although numerous studies are available for the exergy analysis in combustion processes [21,25,29,47,48], no systematic work has been done into the effects of freeboard deflectors on the available exhaust gas power in fixed bed (continuous feed) laboratory scale biomass combustors. The overall aim of this study is to investigate the effects of freeboard deflectors on exergy loss in a laboratory scale fixed bed biomass combustor. For this purpose, experiments have been conducted on a continuous feeding system operated with, and without, a freeboard deflector. The deflector is placed at two different axial positions downstream of the fuel bed. This paper is a follow up on earlier works that look into the effects of freeboard deflectors in (non-reacting) packed beds [45] the fixed bed combustion of biomass [46], and the relevant experimental methodologies [49].
(a)
2. Methodology 2.1. Experimental methodology Experiments are conducted on a laboratory scale fixed bed continuous feeding combustor. Fig. 2a provides a diagram of the data acquisition and control installation. Fig. 2b also depicts the internal detail and four main zones: I- Plenum as well as II- Primary air, IIISecondary air, and IV- Exhaust sections. The overall height of the combustor is 1500 mm, with a 120 mm 120 mm square cross section and is made of high temperature resistant 310 stainless steel (6 mm thick). The exterior walls of the chamber are exposed to ambient conditions (i.e., uninsulated) so as to prevent overheating [15,50–52]. The operational (thermal) power of the combustor largely depends on the primary air flow rate (Qp), and is estimated to be around 15 kW. Both the primary and secondary air (Qs), which is induced further downstream, are supplied by two variable frequency drive centrifugal fans. Flow rates are measured with two mass air flow sensors. To maintain constant air flow rates the fan drives are coupled to a digital controller. Primary air is introduced from beneath through plenum and the grate. The secondary air section is placed at 300 mm above the grate and consists of 44 holes which are uniformly distributed along a single row and centred alongside the inner perimeter of this section.
(b)
Fig. 2. (a) Schematic of experiment setup; (b) CAD generated cross section of the combustor showing the internal future and deflector positions.
65
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Temperature measurements are made by fifteen thermocouples (Type N) coated by a Nicrobel protective metallic sheath. Each of these is 300 mm in length with outer diameter of 3 mm. The operating temperature range of thermocouples is from 270 °C to 1300 °C, at a resolution of 0.15 °C [53]. To resolve the temperature profiles, thermocouple probes (TC0-TC14) penetrate through access ports located at various axial positions along the combustor side wall. To ascertain radial temperature distributions, the thermocouples can be positioned radially by sliding them through each access point. Thermocouples TC0 to TC6 are used to resolve the downstream temperature profiles whilst thermocouples TC7 to TC14 determine upstream temperatures (where only Qp applies). A 16-Channel thermocouple module is used to feed data to an acquisition unit at a sampling rate of 0.2 Hz. In this study, standard cylindrical wood pellets with a 6 mm diameter and characteristic length of 20 mm are used [54]. Table 2 shows the proximate and elemental analysis of the fuel. The pellets are delivered from the hopper by a screw feeder system with a drive motor coupled to a manual on-off digital timer to control the fuel feeding rate. The hopper is suspended from a tensile load cell so as to monitor the mass loss of the fuel. The fuel pellets are fed upwards towards the inclined grate (25 degree gradient) which consists of 120 holes of 4 mm diameter, representing 40% open area (Fig. 2b). The primary air section consists of a detachable viewing port which is used for visual inspection of the combustion process and help a fixed fuel bed height. The exhaust section has an overall height of 1300 mm and located above the secondary air section. The square shaped deflector (95 mm 95 mm) is made of carbon steel and has a thickness of h = 5 mm. The deflectors axial position (H) can be regulated by an adjustable rod (with length of 400 mm) which is connected to a holding flange. The holder flange is installed 300 mm above the secondary air section and has an open area of 90%, whereby the adjusting rod passes through a bush hold by radial spokes which are laser cut into the holder flange. In this study, experiments have been performed for two different deflector positions. The deflector at H = 240 mm is positioned upstream of the secondary air, but between thermocouples TC7 and TC8. Other tests done with the deflector at H = 425 mm mean it is positioned downstream of the secondary air, between thermocouples TC5 and TC6. A continuous sampling gas analyser is used to measure CO, CO2, O2 and NO concentrations, with a suction probe located in the exhaust section at 110 mm above the last thermocouple (TC0). A LabView interface (Ver. 2014) is used to generate a time series of temperatures and emissions data. More details of the combustor and methodologies used in this study are described in previous publications [45,46,49].
not give information on the nature of internal losses and the quality of energy crossing the system boundary [55]. The maximum efficiency of a steady state process is limited by the second law of thermodynamics. The exergy (availability) of a system is defined as the maximum theoretical work that could be done by the system as it interacts with a specified reference environment and reaches thermal and chemical equilibrium [56]. The ensuing parts of this paper will present an exergy analysis for this fixed bed combustor. Complete high temperature oxidation of the biomass fuel occurs through combustion. During this process, the exergy of the biomass is converted mainly into internal energy of the hot flue gases. The chemical composition and the temperature of the gaseous reaction products depends on a number of factors such as the fuel composition, conditions of air supply (Qp, Qs) and heat loss to the surroundings. Fig. 3 shows a schematic diagram of the assumed control volume for the combustion chamber in a counter-current configuration [57]. Table 3 lists the reference environment properties which chosen in this study. In the pellet combustion process the ambient air is the environment. Variations in the ambient air intensive properties influence the system exergy even when the system’s states remain constant. This can be avoided by defining a reference environment with constant properties [29]. The compositions of the reference environment given in Table 3 are based on average concentrations for ambient air. The standard reference temperature and pressure of T0 = 298.17 K and P0 = 1.01325 105Pa have been defined for the reference environment.
Flue gas
CO, H2, CmHn, CO2, H2O, N2
Q Flame front
Fresh fuel
Biomass fuel
2.2. Exergy analysis
Air The first law of thermodynamics can help drive an energy balance for a system and determine heat and work transfer between the system and its environment. However, an energy balance does
Fig. 3. The assumed control volume governing thermal conversion in a countercurrent fixed bed configuration with continuous (fuel) feed.
Table 2 Nominal characteristics of the biomass fuel used [50]. Fuel type
Wood pellet
Formulaa
CH1.49O0.55
Ash analysis (wt.% of dry ash) Na2O MgO 6.22 10.36 a
Dry basis ash free.
Proximate analysis w.b. (%)
LHV (MJ kg1)
q (kg m3)
qP (kg m3)
Porosity
680
1240
0.45
Fe2O3 0.83
ZnO 0.24
Moist.
Vol.
Char
Ash
7.7
67.9
22.6
1.8
18.3
Al2O3 7.74
SiO2 16.47
P2O5 6.98
SO3 4.25
Cl 3.27
K2O 14.29
CaO 28.03
MnO 1.31
66
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Table 3 Reference environment properties [29].
Ambient air
Chemical composition
Mole fraction
N2 O2 H2O (g) CO2 Ar
0.7560 0.2034 0.0312 0.0003 0.0091
where ech p is the specific pellet chemical energy. Specific pellet chemical exergy has a constant value of 20,838,000 (J kg1) and varies proportionally to fuel consumption rate (kg s1) [29]. Carbon monoxide can be reacted with environmental oxygen to produce environmental carbon dioxide.
1 CO þ O2 ! CO2 2
ð7Þ
The chemical exergy of CO can be calculated by [19]:
Exergy consists of potential exergy (Eph), kinetic exergy (Ekn), physical exergy (Ep) and chemical exergy (Ec) [17,58]:
E ¼ Eph þ Ekn þ Ep þ Ec
ð2Þ
In a thermodynamic system such as an atmospheric pressure combustor where flow velocity and elevation changes have negligible effects [59], Ekn and Eph are not typically taken into account. For the remaining terms, Ep indicates the maximum theoretical useful work; therefore it is derived from the temperature and pressure differences between the system and its environment. The chemical non-equilibrium exergy Ec arises from chemical composition changes in the system [17]. Ekn, Eph and Ep can also be grouped together, naming the combination thermomechanical exergy, whilst the chemical non-equilibrium exergy is also named chemical exergy [29]. The thermomechanical exergy of a flow is defined by [58]:
etm ¼ ðh h0 Þ T 0 ðs s0 Þ
ð3Þ
where etm is the specific thermomechanical exergy (J/mol), h is the specific molar enthalpy, s is the specific molar entropy and T is the temperature. The subscript ‘‘0” indicates the reference state. The chemical exergy of a substance depends on its existence in the reference environment. In this research the molecule components of the pellet fuel, CO, N2, O2, H2O(g) and CO2, which participate in the elemental reaction of combustion, have been considered in the exergy analysis. Among these materials only the pellet and CO are not included in the considered environment (Table 3). Since the amount of the other emissions was not significant [46] only CO is taken into account. The general expression for the availability of flue gases is given by [19]:
where R is ideal gas constant, xk is the species mole fraction and x0 is the mole fraction of the element in reference environment. The first underlined term on the right side shows the contribution of temperature to the combustion gases availability, the second underlined term accounts for the chemical exergy (for the substances which are founded in the reference environment) and the third accounts for the pressure being different from P0. A biomass combustion process could be expressed as [29]:
b c b O2 ! aCO2 þ H2 O Ca Hb Oc þ a þ 4 2 2
ð5Þ
The chemical exergy of the pellet fuel is given by:
b c b h h ech ¼ h þ a þ ah p O CO h OðgasÞ p 2 2 4 2 2 2 ðT 0 ;Po Þ b c b sO2 asCO2 sh2 OðgasÞ T 0 sp þ a þ 4 2 2 ðT 0 ;P o Þ 2 3 b c aþ 0 4 2 ðxO2 Þ 5 þ RT 0 ln 4 2 a 0 ðxCO2 Þ ðx0H2 O Þ
1 1 h s ech ¼ h þ h T sh þ s CO O CO2 0 CO O CO2 CO 2 2 2 2 ðT 0 ;P o Þ ðT 0 ;Po Þ 2 3 1 0 2 ðxCO ÞðxO2 Þ 5 þ RT 0 ln 4 x0CO2
ð8Þ
where ech CO is the specific CO pellet chemical exergy. Thermomechanical exergy and chemical exergy can be grouped together, naming the total exergy and expressed as a molar basis as: ch e ¼ ef þ ech p þ eCO
ð9Þ
2.3. Test protocol Before igniting the fuel, biomass pellets are momentarily fed by the screw feeder to the grate and the bed height is set at 10–20 mm below the lowermost thermocouple TC14. After setting the bed height, the primary air is then adjusted and data acquisition started (t = 0 s) using the LabView software interface. With the viewing port removed, a heat gun is used to ignite the top layer of the solid fuel bed. As the flame spreads over top layer of the fuel bed, the viewing port window is reattached and the secondary air regulated. At this stage, the screw feeder driver motor is then once again activated and the mass flow of wood pellets into the combustor controlled by an on-off digital timer controller. The start-up period may take up to 1500 sec (Stage-I in Fig. 4). To self-extinguish the flame at the end of any test run, the screw feeder driver motor is stopped as well as both the primary and secondary supply of air. The shut-down period may take 600 sec (Stage-III in Fig. 4). Fig. 4 presents a time series for the evolution of downstream temperatures (TC0-TC7) and the rate of withdrawing pellets from the hopper for a single test. The steady state time period occurs within the start-up and shut-down (Stages II in Fig. 4). To characterize deflector effects on the exergy in this research, all the combustion parameters are extracted over the steady state time period for each test case. The method which is used to resolve steady state operating conditions can be found in [49]. A MATLAB (Ver. R2012b) script was developed to postprocess the data and calculate the flue gases availability. For exergy analysis of flue gases the average temperatures of TC0-TC2 which are very close to the emission measurement port (Fig. 2) have been used. The routine for calculating the exergy from the test cases presented, was checked against a worked example from literature [19]. Comparisons based on similar temperatures and species (CO) showed that the accuracy between the methods applied in this study and those in the other reference study where between ±2%. 2.4. Uncertainty analysis
ð6Þ
Every measurement is subject to some uncertainty [60]. Measurement errors can come from the measuring instrument such as thermocouple and gas analyser, from the operator, from the environment and from other sources. All experimental uncertainty is due to either random errors or systematic errors [60–62]:
67
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Fig. 4. Time series of downstream thermocouple data and residual fuel in the pellet hopper (Test 54).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
eTotal ¼ e2s þ e2r
ð10Þ
where eTotal is the total uncertainty, es is the systematic error and er is the random error. Systematic errors in experimental observations usually come from the measuring instrument limits (i.e. the accuracy of measurement device). Whilst random errors are statistical fluctuations in the measured data which caused by unknown and unpredictable changes in the experiment. These changes may occur in the measuring instruments or in the environmental conditions [61].
es ¼ er ¼
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Xn
e
2 i¼1 s;i
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi Xn
e
2 i¼1 r;i
ð11Þ ð12Þ
where n is the number of error source. The individual component of the random error in Eq. (12) is calculated as:
er;i
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi PN 2 rs i¼1 ð/i /Þ ¼ ¼ pffiffiffiffi NðN 1Þ N
N X ¼1 / / N i¼1 i
ð13Þ
ð14Þ
is the mean where rs is the standard deviation of a sample mean, / of a measured parameter and N is the number of samples in the repeated observations. 2.4.1. Systematic uncertainties Uncertainties in temperature and emission measurements are mainly associated with the accuracy of the thermocouples and gas analyser. Thermocouples used for measurements are Type N, (make: TC measurement, model: 2I-Nickel-Silicon-Magnesium) and are rated over a temperature range of 270 °C to 1300 °C. Within this range the temperature data acquisition system has an accuracy of 0.4% [53]. A continuous sampling gas analyser (make: Servomex, model: SERVOPRO 4900) is used to monitor fire species (O2, CO, CO2, NO) in the flue gas. Table 4 presents the accuracy of the gas analyser [63]. 2.4.2. Random uncertainty Random errors can be reliably estimated by repeating measurements. Measured data from tests 32, 33 and 34 have been chosen to estimate the random errors (Table 5). These test share same
operating conditions (Qp, Qs) and have similar combustion stoichiometry (0.43 < k < 0.51). More details can be found in Appendix A. 3. Results and discussion The primary air flow rates Qp result in an overall rich combustion stoichiometry (0.208 kprimary 0.719) in the upstream (TC0 to TC7, x = 0 mm to 265 mm). However, once the secondary air is introduced further down at x = 270 mm, the total air flow rates (primary + secondary) lead to lean combustion conditions (1.041 ktotal 3.599). This stoichiometry is more relevant when it comes to analysing the downstream flue gases. The relatively high ratio of secondary air (Qs/Qp) is introduced to prevent particulate (soot) deposition on Sections 2–4 of the combustor and their thermocouples. The experimental data which follows has banded into three ranges of stoichiometry (kprimary) and covers tests with deflectors at two axial distances (H = 240 and 425 mm) as well as baseline tests with no deflector (denoted ‘‘ND”). Table 5 presents the operating conditions of the selected tests. A sample error analysis for temperatures and emissions data has been applied to a limited number of test cases which feature largely similar conditions (Table 5, tests 32, 33 and 34) [10]. Whilst a much wider testing campaign featuring multiple repetitions of similar tests, is warranted so as to present a broader and more representative error analysis, Appendix A presents a summary of the uncertainty analysis applied to these three cases, whereby the likely ±3% variation on exergies has been estimated. Fig. 5 presents the normalized total exergy and CO chemical exergy for all the tests, with and without deflectors. Results are given for thermocouples placed at three radial positions (r = 5, 30 and 60 mm). The secondary horizontal axis shows the kTotal. Data presented is normalized by the peak exergy in each plot, for combustor operation either without a deflector (Fig. 5a) or with deflectors (Fig. 5b). The data clearly indicate the decline in both total exergy and CO exergy with increased k, whether the Table 4 Accuracy of the gas analyser. Gas
Accuracy
O2 CO CO2 NO
0.01% 1% 1% 1%
68
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Table 5 Operating condition of tests. Average fuel flow rate (gr/s)a
Deflector position, H (mm)
Thermocouple position, r (mm)
184 212 240 240 240 246 246 212 240 240 184
0.65 0.583 0.599 0.526 0.503 0.551 0.566 0.440 0.460 0.466 0.437
240 240 240 240 240 ND ND ND 425 ND ND
5 5 60 60 60 60 60 5 30 30 30
60 60 60 60 60 60 60 60 66 60 46 53 66 60 60 60
240 240 240 240 240 240 240 240 246 240 184 212 246 240 240 240
0.413 0.392 0.418 0.427 0.374 0.400 0.390 0.385 0.438 0.363 0.275 0.323 0.423 0.380 0.356 0.374
240 ND 425 425 240 425 ND ND ND 240 240 ND ND 425 240 ND
30 5 5 30 5 60 30 5 5 5 60 5 5 60 5 60
66 46 73 86 66 60 66 66 60 66 53 66 66 60 73 53
264 184 292 344 246 240 246 246 240 246 212 246 246 240 292 212
0.410 0.275 0.409 0.500 0.378 0.340 0.362 0.371 0.329 0.330 0.263 0.340 0.327 0.277 0.352 0.233
425 240 ND ND ND 425 240 240 425 240 240 425 ND ND ND ND
5 60 60 5 5 60 5 5 5 60 60 60 30 60 60 30
kprimary
Test
Q
k = 0.20–0.43
0.208 0.287 0.317 0.357 0.357 0.377 0.378 0.382 0.423 0.423 0.437
59 60 45 31 37 49 5 52 24 13 16
46 53 60 60 60 66 66 53 60 60 46
k = 0.43–0.51
0.439 0.439 0.441 0.447 0.461 0.475 0.475 0.475 0.478 0.489 0.491 0.492 0.494 0.500 0.506 0.509
35 43 30 40 33 25 41 17 54 32 63 22 53 29 34 48
k = 0.51–0.719
0.510 0.537 0.542 0.545 0.554 0.570 0.591 0.592 0.607 0.607 0.610 0.615 0.640 0.691 0.704 0.719
58 55 12 19 21 26 62 61 23 65 64 56 51 42 6 50
P
(l/min)
Q
s
(l/min)
ND: No deflector. a Steady state operation regime.
deflector is present or not. The lower exergy observed at leaner conditions may be attributed to the falling (average) temperature in downstream locations. It can be seen that in tests conducted both with the deflector (Fig. 5b), the data points for both exergies appear tighter (closer) together compared to tests without the deflectors. The significance of this however requires further research. The reasons for this behaviour could be due to the effect of deflectors on the near wall temperatures and CO concentrations [46]. In order to observe the effects of the deflector on the mechanical exergy (which is due to temperature differences over the axial locations in combustor) over kprimary = 0.439–0.509, Fig. 6 presents axial distributions of mechanical exergy for freeboard deflectors placed either upstream of the secondary air at H = 240 mm (Fig. 6b) or downstream of the secondary air at H = 425 mm (Fig. 6c). In this analysis, the flue gas is assumed as (all) air, since the emissions data is acquired only at the gas analyser sample port which is located at the exhaust section of the combustor. The effective temperature is defined as the average of radial temperatures at each axial position. It can be seen that for tests conducted with deflectors both upstream and downstream of the secondary air, the presence of a deflector appears to decrease the mechanical exergy downstream of the secondary air (TC0-TC5, Fig. 6b and c)
when compared with test cases without a deflector (Fig. 6a). However the contribution of the mechanical exergy is minimal towards the total exergy in comparison to the chemical exergy. Deflectors located downstream of the secondary air at H = 425 mm have more impact on upstream mechanical exergy (TC0-TC5, Fig. 6c). It is clear from these results that the magnitude of change in downstream mechanical exergy (TC0-TC0) is sensitive to deflector distance (H). However, the presence of a deflector whether upstream or downstream of the secondary air does not have significant impacts on the upstream mechanical exergy (TC7-TC13, Fig. 6). It is believed the effect on downstream mechanical exergy is due to the impacts of deflector on the temperature profiles in the downstream [46]. To further illustrate the global effects of deflectors on the exergy, Fig. 7 shows the total and CO chemical exergy of the flue gases for several test cases with the same primary air (Qp) and secondary air (Qs), and sharing a narrow range of stoichiometry (kprimary = 0.439–0.509), both with and without deflector. The exergy analysis is based on the emission concentrations at the exhaust section and average temperatures near the gas analyser sampling port (TC0-TC2). It is clear that, using deflectors does not have a significant impact on the total exergy and CO chemical exergy over the range of conditions tested.
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
69
(a) Tests without deflector
(b) Tests with deflectors Fig. 5. Total exergy and CO chemical exergy for tests with and without deflectors, with temperature data at various radial locations.
Fig. 6. Axial mechanical exergy distributions (a) test without deflector; (b) tests with deflectors placed at H = 240 mm; (c) tests with deflectors placed at H = 425 mm (0.43 kprimary 0.51).
Whilst using a deflector generally increases the level of the CO emissions [46], the magnitude of change in CO emissions is not sufficient to appreciably influence the chemical exergy (Fig. 7b). Whilst, the data shows that freebord deflectors affect the axially resolved mechanical exergy along the downstream length of the
combustor (Fig. 6), their effect on the flue gas exergy appears negligible (Fig. 6a). This may be due to the fact that for temperatures up to 208 °C the contribution associated with composition is dominant, whilst above 208 °C the contribution associated with temperature increases with temperature [19]. In this study, the
70
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
Fig. 7. Total and CO chemical exergy for several conditions sharing same Qp and QS, both with and without deflector (0.43 kprimary 0.51).
temperature range of exhaust gas is 180 °C to 220 °C (Fig. 4). As such, whilst freeboard deflectors do appears to influence the axially resolved mechanical exergy, the overall variation of the mechanical exergy changes, both with/without deflectors, is insignificant. 4. Conclusions This study focuses on the effects of freeboard deflectors on the flue gases availability. Experiments were conducted on a continuous feed laboratory scale fixed bed combustor featuring both primary and secondary air, with a freeboard deflector placed at different axial locations. An exergetic analysis was applied to evaluate both the axially resolved mechanical exergy as well as the fume exergetic content of flue gases at the exit plane. A (sample) uncertainty analysis has also been applied. The following conclusions have been drawn as a result of the range of test conditions in this research: Results show that both CO chemical exergy and total exergy decrease when the air-fuel equivalence ratio (k) increases. The decline in CO chemical exergy and total exergy show similar trends for tests both with and without freeboard deflectors. The axially resolved mechanical exergy is affected by the use of freeboard deflectors. The inclusion of a freeboard deflector appears to decrease the mechanical exergy profiles in the downstream (TC0-TC5). This may be due to the heat transfer effects associated with freeboard deflectors but required further investigations. The mechanical exergy is sensitive to the axial position of the deflector (H). By raising the deflector position further above the packed bed, it more significantly affects the mechanical exergy in the downstream. Results reveal the deflectors do not have significant impact on the total exergy and CO chemical exergy of the flue gases at the exhaust section. This may be happen because the contribution associated with mechanical exergy at exhaust section (T < 220 °C) is insignificant. An uncertainty analysis has been applied to a sample number of text cases, whereby the likely ±3% variation on temperature, CO emissions as well as exergies has been estimated. Whilst the results presented herein have been derived over a specific (limited) range of stoichiometry and data set size, the observations made do indicate that more research is needed to ascertain the overall merits of using freeboard deflectors on a wider range of stoichiometric and fixed bed combustor geometries.
Acknowledgements Ms. Araceli Regueiro Pereira is gratefully acknowledged for assisting with some aspects of the data acquisition undertaken at the University de Vigo (Spain). The PhD research project of Mr. Babak Rashidian is made possible by an Edith Cowan University International Postgraduate Research Scholarship (ECU-IPRS).
Appendix A Table A-1 shows the average values of thermocouples and gas species data extracted over the steady state period for the selected tests. The total, systematic and random uncertainty values for the temperatures and emissions are derived using Eqs. (10), (11) and (12), respectively. The results are presented in Table A-2. Table A-3 presents the uncertainty values of the chemical exergy, CO exergy and total exergy.
Table A-1 Steady state values of temperatures and emissions for the selected tests. Test 32
Test 33
Test 34
Qp (l/min) Qs (l/min) Deflector Position, H (mm) Thermocouple position, r (mm)
60 240 240 5
60 240 240 5
60 240 240 5
Temperature (°C)
TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8 TC9 TC10 TC11 TC12 TC13 TC14
229.461 264.932 254.252 273.605 282.109 294.271 314.590 519.970 445.363 460.501 460.939 480.386 555.232 490.465 510.844
239.696 282.338 271.090 278.596 287.914 309.605 347.011 506.943 432.517 432.494 412.158 419.533 509.101 419.734 475.846
198.075 228.245 219.128 227.167 233.783 250.093 268.958 506.456 460.959 477.918 479.803 503.755 568.176 451.352 526.413
Emissions
NO (vpm) O2 (%) CO (%) CO2 (%)
52.848 14.909 1.391 8.696
54.247 14.443 1.461 8.718
50.878 15.048 1.566 8.611
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72 Table A-2 Systematic, random and total uncertainty of measured parameters. Parameter
Systematic uncertainty (±%)
Random uncertainty (±%)
Total uncertainty (±%)
Temperature (°C)
TC0 TC1 TC2 TC3 TC4 TC5 TC6 TC7 TC8 TC9 TC10 TC11 TC12 TC13 TC14
0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4 0.4
4.60 5.03 5.04 5.15 5.23 5.12 5.96 0.71 1.50 2.36 3.65 4.38 2.69 3.68 2.42
4.61 5.05 5.05 5.16 5.24 5.13 5.97 0.81 1.56 2.40 3.67 4.40 2.72 3.70 2.45
Emissions
NO O2 CO CO2
1 0.01 1 1
1.51 1.01 2.82 0.31
1.82 1.01 2.99 1.05
Table A-3 Uncertainty values of chemical exergy, CO exergy and total exergy. Parameter
Systematic uncertainty (±%)
Random uncertainty (±%)
Total uncertainty (±%)
Chemical exergy CO exergy Total exergy
1.00
1.89
2.14
0.4 1.78
1.36 1.58
1.69 2.38
References [1] T. Bridgwater, Biomass for energy, J. Sci. Food Agric. 86 (12) (2006) 1755–1768, http://dx.doi.org/10.1002/jsfa.2605. [2] S. Van Loo, J. Koppejan, The Handbook of Biomass Combustion and Co-Firing, Earthscan, UK, 2008, pp. 38–50. [3] C. Yin, L. Rosendahl, S.K. Kaer, Grate-firing of biomass for heat and power production, Prog. Energy Combust. Sci. 34 (2008) 725–754. [4] C. Yin, L. Rosendahl, S. Kaer, S. Clausen, S. Hvid, T. Hille, Mathematical modeling and experimental study of biomass combustion in a thermal 108 MW grate-fired boiler, Energy Fuels 22 (2) (2008) 1380–1390. [5] Y. Shynya, M. Yukihiko, The Asian biomass handbook, A guide for biomass production and utilization, The Japan Institute of Energy, Tokyo, Japan, 2006, p. 207–222. [6] H. Khodaei, Y.M. Al-Abdeli, F. Guzzomi, G.H. Yeoh, An overview of processes and considerations in the modelling of fixed-bed biomass combustion, Energy 88 (2015) 946–972. [7] C. Wurzenberger, S. Wallner, H. Raupenstrauch, Thermal conversion of biomass: comprehensive reactor and particle modeling, AIChE J. 48 (10) (2002) 2398–2411. [8] M. Katunzi, Biomass conversion in fixed bed experiments, Report number WVT 2006.18, Eindhoven University of Technology, Department of Mechanical Engineering: Netherlands, 2006, p. 1–55. [9] Y.B. Yang, C. Ryu, A. Khor, N.E. Yatesb, V.N. Sharifi, J. Swithenbank, Effect of fuel properties on biomass combustion. Part II. Modelling approach-identification of the controlling factors, Fuel 84 (16) (2005) 2116–2130. [10] B. Rashidian, Deflector effects in fixed bed (biomass) combustors and noncombusting packed beds (Ph.D. thesis), Edith Cowan University, School of Engineering, 2016. [11] C. Ryu, A. Phan, Y.B. Yang, V.N. Sharifi, J. Swithenbank, Ignition and burning rates of segregated waste combustion in packed beds, Waste Manage. 27 (6) (2006) 802–810. [12] T. Nussbaumer, C. Czasch, N. Klippel, J. Johansson, C. Tullin, Primary and secondary measures for the reduction of nitric oxide emissions from biomass combustion, in: Developments in Thermochemical Biomass Conversion, Springer, Netherlands, 1996, p. 1447–1461. [13] C. Ryu, Y.B. Yang, A. Khor, N. Yates, V.N. Sharifi, J. Swithenbank, Effect of fuel properties on biomass combustion: Part 1. Experiments-fuel type, equivalence ratio and particle size, Fuel 85 (2006) 1039–1046.
71
[14] Y.B. Yang, V.N. Sharifi, J. Swithenbank, Effect of air flowrate and fuel moisture on the burning behaviors of biomass and simulated municipal solid wastes in packed beds, Fuel 83 (11) (2004) 1553–1562. [15] H. Wiinikka, R. Gebart, The influence of air distribution rate on particle emissions in fixed bed combustion of biomass, Combust. Sci. Technol. 177 (9) (2005) 1747–1766, http://dx.doi.org/10.1080/00102200590959468. [16] S. Som, A. Datta, Thermodynamic irreversibilities and exergy balance in combustion processes, Prog. Energy Combust. Sci. 34 (3) (2008) 351–376. [17] J. Xiong, H. Zhao, C. Zheng, Exergy analysis of a 600 MWe oxy-combustion pulverized-coal-fired power plant, Energy Fuels 25 (8) (2011) 3854–3864. [18] A. Bejan, M.J. Moran, Thermal Design and Optimization, John Wiley & Sons, 1996, pp. 140–150. [19] M.J. Moran, Availability Analysis: A Guide to Efficient Energy Use, PrenticeHall, New Jersey, 1982, pp. 130–160. [20] Y. Wu, W. Yang, W. Blasiak, Energy and exergy analysis of high temperature agent gasification of biomass, Energies 7 (4) (2014) 2107–2122. [21] K.J. Ptasinski, Thermodynamic efficiency of biomass gasification and biofuels conversion, Biofuels, Bioprod. Biorefin. 2 (3) (2008) 239–253. [22] H. Caliskan, A. Hepbasli, Exergetic cost analysis and sustainability assessment of an internal combustion engine, Int. J. Exergy 8 (3) (2011) 310–324. [23] R. Saidur, J. Ahamed, H. Masjuki, Energy, exergy and economic analysis of industrial boilers, Energy Policy 38 (5) (2010) 2188–2197. [24] M. Hasanuzzaman, N. Rahim, R. Saidurm Analysis of energy, exergy and energy savings of a fire tube boiler, in: 2011 IEEE First Conference on Clean Energy and Technology (CET), IEEE, 2011, pp. 291–296. [25] C. Martín, M.A. Villamañán, C.R. Chamorro, J. Otero, A. Cabanillas, J.J. Segovia, Low-grade coal and biomass co-combustion on fluidized bed: exergy analysis, Energy 31 (2) (2006) 330–344. [26] H. Esen, M. Inalli, M. Esen, K. Pihtili, Energy and exergy analysis of a groundcoupled heat pump system with two horizontal ground heat exchangers, Build. Environ. 42 (10) (2007) 3606–3615. [27] M. Esen, T. Yuksel, Experimental evaluation of using various renewable energy sources for heating a greenhouse, Energy Buildings 65 (2013) 340–351. [28] M.A. Rosen, Energy-and exergy-based comparison of coal-fired and nuclear steam power plants, Exergy Int. J. 1 (3) (2001) 180–192. [29] E. Granada, D. Patiño, J. Collazo, J. Moran, J. Porteiro, Available exhaust gas power in different configurations in a pellet stove plant, Renewable Energy 34 (12) (2009) 2852–2859. [30] W. Dunbar, N. Lior, Sources of combustion irreversibility, Combust. Sci. Technol. 103 (1–6) (1994) 41–61. [31] P. Nicholls, Underfeed Combustion, Effect of Preheat, and Distribution of Ash in Fuel Beds, Bulletin of the United States Bureau of Mines, 1934, p. 378. [32] R. Gort, On the propagation of a reaction front in a packed bed: thermal conversion of municipal waste and biomass (Ph.D. thesis), Universiteit Twente, Enschede, 1995. [33] T. Rogaumea, M. Auzanneau, F. Jabouille, J.C. Goudeau, J.L. Torerob, The effects of different airflows on the formation of pollutants during waste incineration, Fuel 81 (2002) 2277–2288. [34] H. Wiinikka, R. Gebart, Critical parameters for particle emissions in small-scale fixed-bed combustion of wood pellets, Am. Chem. Soc. J. 18 (2004) 898–907. [35] R.P. Van der Lans, L.T. Pedersen, A. Jensen, P. Glarborg, D. Johansen, Modelling and experiments of straw combustion in a grate furnace, Biomass Bioenergy 19 (2000) 199–208. [36] J.J. Saastamoinen, M. Horttanainen, P. Sarkomaa, Ignition wave propagation and release of volatiles in beds of wood particles, Combust. Sci. Technol. 165 (1) (2001) 41–60. [37] J.I. Samuelsson, Conversion of Nitrogen in a Fixed Buring Biofuel Bed Thesis for the degree of licentiate of engineering, Chalmers University of Technology, 2006. [38] A. Weissinger, T. Fleckl, I. Obernberger, In situ FT-IR spectroscopic investigations of species from biomass fuels in a laboratory-scale combustor: the release of nitrogenous species, Combust. Flame 137 (4) (2004) 403–417. [39] N. Woudstra, T. van der Stelt, Exergy analysis of combustion systems, in: Proceedings of the International ECOS Conference, Copenhagen, Denmark, June, 2003. [40] N. Klippel, T. Nussbaumer, Health relevance of particles from wood combustion in comparison to diesel soot, in: 15th European Biomass Conference, Berlin, 2007. [41] B. Kovacevik, Alkali-containing aerosol particles release during biomass combustion and ambient air concentrations (Ph.D. thesis), University of Gothenburg, Department of Chemistry, Atmospheric Science, 2009. [42] K. Porle, N. Klippel, O. Riccius, I.I. Kauppinen, T. Lind, Full scale ESP performance after PC-boilers firing low sulfur coals, in: Proceedings of EPRI/ DOE International Conference of Managing Hazardous and Particulate Air Pollutants, Toronto, Canada, 1997. [43] L. Salaver, T. Lee, C.D. Cooper, C. Yu Wu, J. Donnelly, Air Pollution Emission Control Devices for Stationary Sources, Air & Waste Management Association (A&WMA), 2007. [44] R. Buczyn´ski, R. Weber, A. Szle˛k, Innovative design solutions for small-scale domestic boilers: Combustion improvements using a CFD-based mathematical model, J. Energy Inst. 88 (1) (2015) 53–63. [45] B. Rashidian, Y.M. Al-Abdeli, G.H. Yeoh, F.G. Guzzomi, Effect of freeboard deflectors on the temperature distribution in packed beds, Appl. Therm. Eng. 89 (2015) 134–143.
72
B. Rashidian, Y.M. Al-Abdeli / Applied Thermal Engineering 118 (2017) 62–72
[46] B. Rashidian, Y.M. Al-Abdeli, D. Patiño, F.G. Guzzomi, G.H. Yeoh, Effect of freeboard deflectors in the fixed bed combustion of biomass, Appl. Therm. Eng. 103 (2016) 543–552. [47] A. Abuadala, I. Dincer, G. Naterer, Exergy analysis of hydrogen production from biomass gasification, Int. J. Hydrogen Energy 35 (10) (2010) 4981– 4990. [48] T. Nussbaumer, M. Oser, Evaluation of Biomass Combustion Based Energy Systems by Cumulative Energy Demand and Energy Yield Coefficient, International Energy Agency and Swiss Federal Office of Energy, Zurich, 2004, January. [49] B. Rashidian, Y.M. Al-Abdeli, G.H. Yeoh, D. Patiño, F. Guzzomi, Methodologies for processing fixed bed combustor data, Combust. Sci. Technol. 189 (1) (2017) 79–102. [50] J. Porteiro, E. Granada, J. Collazo, D. Patiño, J. Morán, Experimental analysis of ignition front of several biomass fuel in a fixed-bed combuster, Fuel 89 (1) (2010) 26–35. [51] L. Febrero, E. Granada, D. Patiño, P. Eguía, A. Regueiro, A comparative study of fouling and bottom ash from woody biomass combustion in a fixed-bed smallscale boiler and evaluation of the analytical techniques used, Sustainability 7 (5) (2015) 5819–5837. [52] L. Febrero, E. Granada, A. Regueiro, J.L. Míguez, Influence of combustion parameters on fouling composition after wood pellet burning in a lab-scale low-power boiler, Energies 8 (9) (2015) 9794–9816. [53] NI 9213 Operating instructions and specifications, N.I. Corporation, Editor, Austin, USA, 2009.
[54] J. Porteiro, D. Patiño, J. Moran, E. Granada, Study of a fixed-bed biomass combustor: Influential parameters on ignition front propagation using parametric analysis, Energy Fuels 24 (7) (2010) 3890–3897, http://dx.doi. org/10.1021/ef100422y. [55] R. Prommas, P. Rattanadecho, D. Cholaseuk, Energy and exergy analyses in drying process of porous media using hot air, Int. Commun. Heat Mass Transfer 37 (4) (2010) 372–378. [56] I. Dincer, Y.A. Cengel, Energy, entropy and exergy concepts and their roles in thermal engineering, Entropy 3 (3) (2001) 116–149. [57] H. Thunman, B. Leckner, Co-current and counter-current fixed bed combustion of biofuel-a comparison, Fuel 82 (2003) 275–283. [58] N. Shigaki, T. Akiyama, F. Tsukihashi, Exergy analysis of steel production processes, Mater. Trans. 43 (3) (2002) 379–384. [59] P. Ahmadi, I. Dincer, Exergoenvironmental analysis and optimization of a cogeneration plant system using Multimodal Genetic Algorithm (MGA), Energy 35 (12) (2010) 5161–5172. [60] Z.U. Ahmed, An experimental and numerical study of surface interactions in turbulent swirling jets (Ph.D. thesis), Edith Cowan University, School of Engineering, 2016. [61] R.J. Moffat, Describing the uncertainties in experimental results, Exp. Thermal Fluid Sci. 1 (1) (1988) 3–17. [62] D.C. Montgomery, G.C. Runger, Applied Statistics and Probability for Engineers, fifth ed., John Wiley & Sons, USA, 2010, pp. 370–380. [63] ServoPro Configuration and operation 4000 Series, S.G. Limited, Editor, East Sussex, UK, 2011.