Available online at www.sciencedirect.

com
Available online at www.sciencedirect.com
ScienceDirect
ScienceDirect
Energyonline
Available
Available Procedia 00
onlineatat (2018) 000–000
www.sciencedirect.com
www.sciencedirect.com
Energy Procedia 00 (2018) 000–000 www.elsevier.com/locate/procedia

ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy
EnergyProcedia
Procedia158 (2019) 000–000
00 (2017) 898–903
www.elsevier.com/locate/procedia
10th International Conference on Applied Energy (ICAE2018), 22-25 August 2018, Hong Kong,
10th International Conference on Applied Energy
China(ICAE2018), 22-25 August 2018, Hong Kong,
China
Effect of different operating conditions on the performance of a 32
Effect of different operatingSymposium
The 15th International conditions on the
on District performance
Heating and Cooling of a 32

Assessing the MW woodchip-fired

feasibility of using grate boiler
the heat demand-outdoor
MW woodchip-fired grate boiler
temperature a,bfunction for a long-term district heat demand forecast
Yaojie Tu *, Wenming Yang , Keng Boon Siah , Subbaiah Prabakaran
b c c

I.Yaojie Tu *,*,A.
Wenming , P.Yang , Keng Boon Siah ., B.,Singapore
Subbaiah cPrabakaran
a,b
a,b,c b a c c
Andrić
Sembcorp-NUS
a PinaaLaboratory,
Corporate Ferrão , J.University
National Fournier b
of Singapore, Lacarrière O. Le Correc
117576,, Singapore
b
Department aof Mechanical Engineering, Faculty of Engineering, NationalofUniversity of Singapore, Singapore 117576, Singapore
a
IN+ Center for Sembcorp-NUS Corporate
Innovation, Technology Laboratory,
and National
Policy Research University
- Instituto Singapore,
Superior Técnico,Singapore 117576,
Av. Rovisco Pais 1,Singapore
1049-001 Lisbon, Portugal
b
Department of Mechanical
b
c
Sembcorp Faculty
Engineering, Industries Ltd, 30 Hill Street,
of Engineering, #05-04,
National 179360,
University Singapore Singapore 117576, Singapore
of78520
Singapore,
Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, Limay, France
c
c
Sembcorp Industries
Département Systèmes Énergétiques Ltd, 30 Hill- Street,
et Environnement #05-04, 179360,
IMT Atlantique, Singapore
4 rue Alfred Kastler, 44300 Nantes, France

Abstract
Abstract
This paper presents a numerical study on the effect of different operating conditions (excess air coefficient, in-
Abstract
This pressure
furnace paper presents
and airaratio
numerical
in zonestudy
5) onon the effect grate-firing
a woodchip of different boiler.
operating conditions
First, (excess
the numerical air coefficient,
modeling approachin-is
furnace
validatedpressure and
with networks
District heating air ratio
experimental in zone 5) on
measurement
are commonly a woodchip
addressed grate-firing
of inin-furnace
the literature boiler.
temperature First,
as one of theasmost the numerical
welleffective modeling
as exhausted approach
gasdecreasing
solutions for emissions.is
the
validated with experimental
emissions from measurement
the building of
sector. in-furnace
These systems temperature
require high as well
investments as
Subsequently, parametric studies are carried out using the validated numerical modeling approach. The modeling
greenhouse gas whichexhausted
are returnedgas emissions.
through the heat
Subsequently,
sales. Due
results indicate parametric
to the changed
that, studies
the boiler are carried
climateoverall
conditions out building
and
efficiency using
can be the validatedpolicies,
renovation
improved numerical
through heat modeling
demandthe
adjusting inapproach.of The
the future
values modeling
could decrease,
the operating
prolonging
results the
indicate investment
that, the return
boiler period.
overall efficiency can be improved through adjusting the
conditions of interest. It is suggested to improve the boiler efficiency through reducing excess air coefficient, values of the operating
The main scope
conditions
increasing of thispressure
of interest.
in-furnace paper
It isis suggested
toand
assess thetofeasibility
enhancing improve ofthe
using
ash cooling inthe
boilerheat
zone 5.demand –through
efficiency outdoor temperature functionair
reducing excess forcoefficient,
heat demand
forecast. The
increasing district of
in-furnace Alvalade,
pressure andlocated
enhancingin Lisbon (Portugal),
ash cooling was5.used as a case study. The district is consisted of 665
in zone
buildings that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district
Copyright © 2018 Elsevier Ltd. All rights reserved.
©renovation scenariosPublished
2019 The Authors. were developed (shallow,
by Elsevier Ltd. intermediate, deep). To estimate the error, obtained heat demand values were
Selection and
Copyright peer-review
© 2018 under
Elsevier Ltd. Allresponsibility
rights reserved.of the scientific committee of the 10th International Conference on Applied
This is an open
compared withaccess
resultsarticle
from aunder the CC
dynamic heatBY-NC-ND
demand model, license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
previously developed and validated by the authors.
Energy
Selection(ICAE2018).
Peer-review and peer-review underofresponsibility of the scientific committee
– Theofof theInternational
10 International
th Conference on Applied
The results under responsibility
showed that when only the scientific
weather committee
change of ICAE2018
is considered, the margin 10th Conference
error could be acceptable onsome
for Applied Energy.
applications
Energy (ICAE2018).
(the error in annual demand was lower than 20% for all weather
Keywords: Grate boiler, operating parameters, boiler efficiency, CFD modeling
scenarios considered). However, after introducing renovation
scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered).
Keywords: Grate boiler, operating parameters, boiler efficiency, CFD modeling
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the
decrease in the number of heating hours of 22-139h during the heating season (depending on the combination of weather and
renovation scenarios considered). On the other hand, function intercept increased for 7.8-12.7% per decade (depending on the
coupled scenarios). The values suggested could be used to modify the function parameters for the scenarios considered, and
improve the accuracy of heat demand estimations.

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
Cooling.
* Corresponding author.
* E-mail mpeywm@nus.edu.sg (Yang Wenming)
address:author.
Corresponding
Keywords: Heat demand;
E-mail address: Forecast; Climate
mpeywm@nus.edu.sg change
(Yang Wenming)
1876-6102 Copyright © 2018 Elsevier Ltd. All rights reserved.
Selection and
1876-6102 peer-review
Copyright under
© 2018 responsibility
Elsevier Ltd. All of the scientific
rights reserved. committee of the 10 International Conference on Applied Energy (ICAE2018).
th

Selection and peer-review under responsibility of the scientific committee of the 10th International Conference on Applied Energy (ICAE2018).
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2019 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Peer-review under responsibility of the scientific committee of ICAE2018 – The 10th International Conference on Applied Energy.
10.1016/j.egypro.2019.01.228
 Yaojie Tu et al. / Energy Procedia 158 (2019) 898–903 899
2 Author name / Energy Procedia 00 (2018) 000–000

1. Introduction

Biomass is a renewable energy resource and has the advantage of being CO2 neutral with comparison to fossil
fuels. At present, the most commonly used way to utilize biomass for heat and power production is through direct
combustion in a grate boiler [1]. Recently, the interest of optimizing the performance of grate boilers has grown fast,
mainly focusing on improving boiler thermal efficiency and reducing pollutant emissions. For such purpose,
computer-aided modeling techniques have become a cost-effective tool with comparison to practical experimental
tests. In particular, the computational fluid dynamics (CFD) modeling is among the techniques for visualizing the
details of the combustion inside a grate boiler. With such tool, one is able to predict the effect of various parameters
on the boiler performance, for instance, air staging [2], fuel property [3] and burner arrangement [4] and so on.
Frankly speaking, the modification of the above-mentioned parameters essentially requires the retrofit of the
boiler construction, which is always not desired by the plant. On the other hand, limited studies have been dedicated
to reveal the effect of operating conditions (such as excess air coefficient, in-furnace pressure and air distribution on
the grate, etc.) in the open literature on grate boilers. In this study, an attempt will be made to figure out the
influence of several operating parameters of interest on the combustion performance of a woodchip fired grate boiler
using CFD modeling approach. In particular, the focus of the study will be placed on the difference in temperature
distribution as well as the improvement of the boiler overall efficiency while changing the operating conditions.

2. Methodology

2.1. Boiler information

Fig. 1 shows the schematic diagram of the 32 MW grate boiler used in this study. The boiler can be generally
divided into three regions, i.e. the bottom grate combustion region, the upper freeboard combustion region and the
downstream heat exchange region (including evaporator (EVA), super heater (SH) and economizer (ECO)). Under
the design condition, the ratio between primary air (PA) and secondary air (SA) is 42%:58%. PA is sent to the
furnace through five evenly distributed zones on the grate, while SA is delivered through four rows of pipes located
oppositely on the right and left walls.

Fig. 1. Schematic of the woodchip grate-firing boiler.

2.2. Modeling approach

The modeling of the combustion process inside a grate boiler consists of two procedures, namely the biomass
thermochemical conversion on the moving grate and the gaseous species combustion in the upper freeboard. The
900 Yaojie Tu et al. / Energy Procedia 158 (2019) 898–903
Author name / Energy Procedia 00 (2018) 000–000 3

two procedures are strongly coupled by heat and mass exchange. The information (mass flow rate, species
concentrations and temperature) of the emitted gas from the fuel bed is used as boundary condition for the upper
freeboard modeling. For the fuel conversion on the grate, the equilibrium calculation method (ECM) is adopted, the
details of which can be found in [5]. It prescribes the fuel conversion ratios of different processes (evaporation,
devolatilization, char burnout and ash cooling) as a function of the grate location based on both practical operating
experience and experimental measurement. The upper freeboard modelling is conducted using the commercial CFD
package of Fluent, which solves equations of mass and heat transfer as well as chemical reaction in computational
domain. The detailed description of the CFD models can be referred to our previous work [4].

3. Modeling validation

The modeling method has been validated against experimental data of temperature profile and exhausted gas
emissions for thirteen cases. Due to the limitation of the paper length, only one validation result is given in Fig. 2.
Specifically, Fig. 2a shows the locations of the nine thermocouples, and Fig. 2b and Fig. 2c show the comparison of
temperature profiles and exhausted gas emissions between CFD prediction and experimental measurement,
respectively. The comparisons imply that the present modeling method is capable of reproducing the temperature
distribution as well as final gas emission of the woodchip-fired grate boiler. Therefore, in the following parametric
studies, the same modeling method is employed.
24 600

1500 Experiment
CFD modeling
1400 Plant experiment 20 500
1300 CFD modeling
Dry species concentration

Dry species concentration

1200
Gas temperature (K)

16 400

(NO & SO2 in ppm)

(O2 & CO2 in %)
1100
1000
900 12 300

800
700
8 200
600
500
4 100
400
300
0 1 2 3 4 5 6 7 8 9 10
0 0
Measuring point ID O2 CO2 NO SO2

(a) Sampling points (b) Validation of temperature (c) Validation of exhausted gas composition

Fig. 2. Validation results of the CFD modeling method.

4. Results and discussion

4.1. Boiler heat balance analysis

The overall boiler efficiency (η) is defined as the ratio between effective heat absorption (Qeffective) and the total
heat input (Qtotal). The heat loss comes from five major parts, i.e. bottom ash heat loss (Q1), incomplete combustion
in flue gas (Q2), heat exchange through boiler exterior wall (Q3), unburnt carbon in bottom ash (Q4) and stack loss
(Q5). A preliminary analysis of the present boiler heat balance has been made, and the proportions of each heat loss
are shown in Fig. 3. Q1 ~ Q5 are estimated to be 1%, 0.5%, 1%, 0.5% and 7%, respectively, leaving the overall
boiler efficiency approximating to be about 90%. As can be seen, it would be more promising to deal with Q1, Q3
and Q5. However, reducing Q3 means that one needs to reduce the combustion temperature inside the furnace,
which is not desired for the boiler. Hence, we will try to reduce the heat loss through minimizing Q1 and Q5.
 Yaojie Tu et al. / Energy Procedia 158 (2019) 898–903 901
4 Author name / Energy Procedia 00 (2018) 000–000

Fig. 3. Heat balance for the present grate-firing boiler.

4.2. Effect of different operating parameters on boiler performance

4.2.1. Excess air coefficient (λ)

To ensure the complete combustion of fuel, more air is delivered into the furnace. However, the excess air not
only moderates the combustion temperature, but also increases the flue gas amount. Therefore, considering a smaller
λ would be expected to improve η. While its influence on combustion should be also addressed. Hence, five cases
are designated with λ being 1.2, 1.3, 1.4, 1.5 and 1.6, respectively.
Fig. 4 shows the spatial flame kernel (indicated by 1400 K temperature iso-surface) distributions for the five
cases. It is clear that the combustion of volatiles above the grate is intensified by gradually increasing λ. Meanwhile,
the flame kernel shifts rightwards and becomes shortened as λ increases, implying a more concentrated flame and
higher temperature peak. The rightwards shifted flame would also give rise to the ash deposition problem on the rear
wall because of the high temperature and enhanced impinging possibility of the particles released from the fuel bed.

λ=1.2 λ=1.3 λ=1.4 λ=1.5 λ=1.6

Fig. 4. Spatial distribution of flame kernel under different λ.

Fig. 5a and Fig. 5b show the variations of boiler stack temperature and η under different λ conditions. As λ
increases, the boiler stack temperature gradually climbs up. Specifically, the boiler stack temperature will be
increased by about 6 K when λ is enlarged from 1.2 to 1.6. This is mainly caused by the shorter residence time of
flue gas within the furnace owing to the larger gas amount as λ is increased. Since the specific heat of the flue gas
shows a minor difference among the five cases, the stack loss increases as a result of higher stack temperature,
which subsequently leads to the lower η as indicated by Fig. 5b. In particular, when λ increases from 1.2 to 1.6, η
will be reduced by about 1.91%.
460 100
(a) (b)
98
455
96
Stack temperature (K)

Overall efficiency (%)

94
450
92

445 90

88
440
86

84
435
82

430 80
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.1 1.2 1.3 1.4 1.5 1.6 1.7

Excess air coefficient () Excess air coefficient ()

Fig. 5. Effect of λ on (a) stack temperature and (b) boiler efficiency.

4.2.2. In-furnace pressure (P)

For grate boilers, negative P is always kept to prevent hot gas from coming out through openings. However, high
P can affect η by increasing the air infiltration. It means more cold air will enter the boiler furnace chamber,
resulting in lower combustion temperature and higher stack heat loss. The current P measured at zone 5 is about -
150 pa. To reveal the influence of P on boiler combustion, in total five cases are modeled with P at zone 5 being -
100 pa, -125 pa, -150 pa, -175 pa and -200 pa, respectively. Fig. 6 shows the spatial flame kernel distribution in the
boiler for the five cases. It shows that the change of P will not significantly influence the combustion in the upper
furnace chamber. However, the combustion of volatiles above zone 1 can be strengthened.
902 Yaojie Tu et al. / Energy Procedia 158 (2019) 898–903
Author name / Energy Procedia 00 (2018) 000–000 5

p=-100 pa p=-125 pa p=-150 pa p=-175 pa p=-200 pa

Fig. 6. Spatial distribution of flame kernel under different P conditions.

Fig. 7a and Fig. 7b plot the variations of boiler stack temperature and η against P, respectively. It is found that the
stack temperature can be slightly lowered by reducing P. This is mainly because of the smaller gas amount under
lower P conditions. In particular, the gas residence time will be longer and the heat transfer will be enhanced.
Moreover, according to Fig. 7b, η presents a rising trend when gradually reducing P. Specifically, when P is reduced
from -200 pa to -100 pa, η could be improved by nearly 0.97%, which is mainly due to smaller stack heat loss.
460 94
(a) (b)
455
Stack temperature (K)

92
450 Overall efficiency (%)

445 90

440
88
435

430 86
-200 -180 -160 -140 -120 -100 -200 -180 -160 -140 -120 -100
Pressure (pa) Pressure (pa)

Fig. 7. Effect of P on (a) stack temperature and (b) boiler efficiency.

4.2.3. Air ratio in zone 5 (AR5)

The bottom ash heat loss (Q1) can reach up to 3% when burning high ash content fuels. To mitigate this part, it is
expected to improve the ash cooling process by adding more air in zone 5. Currently, the PA percentages in zone 4
and zone 5 are 20% and 10%, respectively. Three cases are modelled by adjusting the air ratio in zone 5 (AR5) to be
10%, 14% and 18%, respectively. Note that, the excess air in zone 5 is taken from zone 4 in order to maintain the
same overall excess air coefficient.
In Fig. 8, the spatial flame kernel distribution of the three studied cases. Both the shape and the size of the flame
kernel are similar to each. It indicates that increasing AR5 from 10% to 18% will have an insignificant impact on the
combustion process in the upper furnace chamber. In particular, the stack temperature for the three cases are 447.62
K, 447.77 K and 447.89 K, respectively, which means the heat transfer in the lower parts (EVA, SH and ECO) is not
affected obviously.

AR5=10% AR5=14% AR5=18%

Fig. 8. Spatial distribution of flame kernel under different AR5 conditions.

 Yaojie Tu et al. / Energy Procedia 158 (2019) 898–903 903
6 Author name / Energy Procedia 00 (2018) 000–000

800 92 1.0
(a) (b) (c)
750
0.76%
Bottom ash temperature (K)

0.8

Efficiency improvement (%)

700 91

Overall efficiency (%)

650
0.6
600 90

550 0.4
0.29%
500 89

450 0.2

0.03%
400 88
10 14 18 10 14 18 0.0
PA percentage in zone 5 (%) PA percentage in zone 5 (%) HQF MQF LQF

Fig. 9. Effect of AR5 on (a) bottom ash temperature, (b) boiler efficiency and (c) efficiency improvement for different fuels.

Owing to the enhanced ash cooling, the bottom ash temperature is found to be reduced by increasing AR5
according to Fig. 9a. Specifically, the bottom ash temperature is reduced by about 120 K when increasing AR5 from
10% to 18%. Consequently, η is improved by 0.29% when using a fuel with an ash content of 21.28% according to
Fig. 9b. In addition, in order to further reveal the potential of improving η by increasing AR5 for different types of
fuels, three categories are classified according to the ash content in fuel. In specific, high quality fuel (HQF) is
defined as the fuel type which has an ash content less than 10%, low quality fuel (LQF) is defined as the fuel type
which has an ash content larger than 30%, and medium quality fuel (MQF) is the fuel type which has an ash content
between 10% and 30%. According to Fig. 9c, the efficiency improvement can be as high as 0.76% for the LQF.
Therefore, it is suggested to increase AR5 when burning a high ash content fuel.

5. Conclusion

The influence of three operating conditions on the performance of a woodchip-fired grate boiler is numerically
examined. The numerical modeling approach has been validated against measured data of temperature and gas
emissions. The parametric studies show that the boiler overall efficiency can be improved by lowering boiler excess
air coefficient, raising in-furnace pressure and increasing air ratio in zone 5. The excess air coefficient is the most
sensitive parameter among them, while increasing air ratio in zone 5 is very promising for high ash content fuels.
However, it is still worth pointing out that although the present study could serve as guidance for practical operation,
the optimal conditions should consider the final CO emission as it might increase in smaller λ conditions. This will
be studied by experimental tests in our next step.

Acknowledgements

The authors acknowledge the financial support by the National Research Foundation of Singapore, Sembcorp
Industries Ltd and National University of Singapore under the Sembcorp-NUS Corporate Laboratory.

References

[1] Yin, Chungen, Lasse A. Rosendahl, and Søren K. Kær. (2008). "Grate-firing of biomass for heat and power production." Progress in Energy
and Combustion Science 34.6 (2008): 725-754.
[2] Staiger, B., S. Unterberger, R. Berger, and Klaus RG Hein. "Development of an air staging technology to reduce NOx emissions in grate fired
boilers." Energy 30.8 (2005): 1429-1438.
[3] Yang, Yao Bin, Robert Newman, Vida Sharifi, Jim Swithenbank, and John Ariss. (2007). "Mathematical modelling of straw combustion in a
38MWe power plant furnace and effect of operating conditions." Fuel 86.1 (2007): 129-142.
[4] Tu, Yaojie, Anqi Zhou, Mingchen Xu, Wenming Yang, Keng Boon Siah, and Prabakaran Subbaiah. "NOX reduction in a 40 t/h biomass fired
grate boiler using internal flue gas recirculation technology." Applied Energy 220 (2018): 962-973.
[5] Costa, M., Nicola Massarotti, V. Indrizzi, Boštjan Rajh, Chungen Yin, and Niko Samec. "Engineering bed models for solid fuel conversion
process in grate-fired boilers." Energy 77 (2014): 244-253.