Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

Modeling and Simulation of Interconnected CFB-boiler and Fast Pyrolysis

Processes Control Design Case
Yrj Majanne* Jani Laine**
Jyri Kaivosoja*** Petri Kykk **

*Tampere University of Technology, Department of Automation Science and Engineering, P.O. Box 692,
FI-33101 Tampere, Finland (e-mail:yrjo.majanne@tut.fi)
** Metso Power, P.O.Box 109, FI-33101 Tampere, Finland (e-mails:
Jani.Laine@metso.com Petri.Koykka@metso.com )
*** Metso Automation, P.O.Box 237, FI-33101, Tampere, Finland
(e-mail: Jyri.Kaivosoja@metso.com)
Abstract: The fast pyrolysis process is developed to produce liquid bio oil from biomass. In the discussed
process the required thermal energy is extracted from a circulating fluidized bed boiler (CFB) as hot sand
fed into the pyrolysis reactor. Residual char and uncondensed gases from the pyrolysis process are fed
back and combusted in the boiler. Thus, the pyrolysis and the boiler processes are interconnected together
both via energy and mass balances. Dynamic simulation is used to analyze how start ups, stops, and
disturbances in the pyrolysis process effect on the operation of the boiler. According to the results of the
analysis the boiler control system is modified to be able to stabilize the operation of the boiler with the
interconnected pyrolysis process. The trouble free operation of the boiler must be guaranteed during all
possible situations of the pyrolysis process. This project is a part of a R&D project by Metso Power
where fast pyrolysis process is developed and commercialized as a new liquid bio fuel production process
which can be integrated e.g. as a part of a paper mill.
Keywords: Dynamic simulation, CFB boilers, Fast Pyrolysis, Interconnected systems, Boiler control.

1. INTRODUCTION
Fuels produced from renewable energy sources are seen as a
part of the solution to reduce CO2 emissions and the
replacement of fossil fuels. Pyrolysis is a promising process
for conversion of solid biomass to liquid bio fuels.
This paper introduces an integrated process concept, where
the pyrolysis process is connected together with a fluidized
bed boiler. The concept is developed by Metso Corporation
and Technical Research Centre of Finland, VTT. Heat
required for the pyrolysis process is extracted as hot sand
from the fluidized bed boiler. The charcoal remained from
the pyrolyzed biomass and non-condensable pyrolysis gases
are fed back to the boiler to be combusted. Thus, both the
energy and the mass balances of the boiler and the pyrolysis
processes are connected with each other.
In the pyrolysis process the volatile components are extracted
from biomass by heating up the solid matter in an oxygen
free atmosphere. After the extraction the gaseous pyrolysis
product is condensed into a liquid phase. This pyrolysis oil
can be used directly to replace heavy fuel oil in industrial
processes or as a raw material for upgrading processes to
synthesize new hydrocarbon compounds, (Sipil et al., 2007),
(Calabria et al., 2005). Research and development of
pyrolysis process is surveyed by Bridgwater (1999) and
Bridgwater & Peacocke (2000).
Modelling of the pyrolysis process is based on the first
principal models of heat and mass transfer and experimental
Copyright by the
International Federation of Automatic Control (IFAC)

results about the reaction kinetics. The applied heat transfer

model is based on the model presented in Saastamoinen,
2006. The pyrolyzer model used in the study is based on the
reactor model developed in a research project with Metso
Power described in (Kaivosoja, 2008) and (Kaivosoja and
Majanne, 2009).
There is a lot of research work going on about the modelling
of combustion of different type of fuels in different type of
combustors, e.g. (Scala & Palatino, 2002) and (Galgano et al.
2005). However, the combustion models developed there are
seldom connected with the dynamic heat transfer models for
water-steam cycle of a boiler. In the dynamic boiler models
the combustion process and phenomena in the furnace have
often been left for a minor consideration. The heat power
released in the furnace is often presented as a simple low
order transfer function describing the dynamics caused by the
fuel transportation, combustion speed and heat storage of
steely heat exchangers bodies. With this kind of approach it is
not possible to represent the interconnections between the
boiler and the pyrolysis processes, where mass and energy
are transported between the two processes effecting on the
fuel power and combustion parameters in the furnace.
The boiler model used in the study is based on the circulating
fluidized bed boiler model developed in a research project
with Metso Power, (Kykk, 2007) and (Majanne and
Kykk, 2009). The simulation models are built and run in
Matlab/Simulink environment. A moderately complicated
combustion model is connected with the water-steam cycle

8283

Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

model resulting a dynamic boiler model taking account also

physical phenomena of the fluidized bed combustion.
The topic of this paper is a simulation study about the
dynamic interactions between the two interconnected
processes. The final goal is to use the analysis results for
process design in order to decouple these two processes from
each other and also to develop process control structures for
improved disturbance rejection.
The structure of this paper is as follows: Section 2 presents
the structure and the operation of the integrated pyrolysis
process. Section 3 presents the applied boiler control system
and the simulation results. Section 4 concludes the paper.
2. INTEGRATED PYROLYSIS PROCESS
The structure of the pyrolysis process integrated with a
circulating fluidized bed (CFB) boiler is shown in Fig. 1.
Combustion gas

Cyclone

Non-condensed
pyrolysis gas

Cooler

From the cyclone hot pyrolysis gases flow to the condenser,

where gas is quenched by cooled pyrolysis oil spray.
Condensable components of the gas condense into liquid
pyrolysis oil. Non-condensable gases are recirculated as an
oxygen free fluidization gas for the reactor. Excessive gas is
blown to the CFB boiler to be combusted.
The structure of the dynamic simulation model is depicted in
fig. 2. Pyrolysis reactor has been divided into four blocks.
The structure of each reactor block is identical. Symbols
inside the elements represent the state variables (see Table 1.)
of the dynamic model.

Condensate
pump

Pyrolysis oil

nd
Sa

Non-condensed pyrolysis gas

Loop
seal

Pyrolysis reactor

Loop
seal

Condenser

Cyclone

Circulating Fluidized Bed Boiler

Boiler
Fuel

The best bio-oil yield is obtained with a high heating rate at

reaction temperatures around 500 C (app. 1 sec.) and with
short vapour residence time before condensation (typically
less than 2 seconds). This type of process is called a fast
pyrolysis process. The characteristic features of the fast
pyrolysis process usually require small particle size of
biomass and rapid condensing of pyrolysis vapours.
On the top of the reactor solids (sand and charcoal) are
separated from gases in a cyclone. From the cyclone solids
flow through a loop seal to the furnace of the CFB boiler. In
the furnace the char coal is combusted and sand and ash are
mixed with the boiler bed.

Pyrolysis gas

,
Char
Sand

Heat required for the pyrolysis reactions is introduced to the

reactor as hot sand extracted from the fluidized bed boiler.
Besides carrying heat power to the process, hot sand particles
also improve the heat transfer efficiency from the reactor
atmosphere to the biomass. Hot sand and biomass are fed into
the reactor together with an inert fluidization gas at the
bottom of the reactor. During the turbulent pneumatic
transportation in the reactor biomass decomposes into four
main components: condensable organic gas, non-condensable
gas, water, and charcoal.

Primary
air

Pyrolysis
reactor model

Aeration
gas
Biomass
Fluidization
gas

xi ,k
Tn ,k
Fluidization gas
compressor

k=4

m tot
xi ,4
Tn ,4

Cyclone and
loopseal model

m gas

Tgas
Tloopseal

xi ( gas )
Tgas

m solid
xi ,k
Tn ,k

Fig. 1 The structure of the integrated pyrolysis process.

(Kaivosoja and Majanne, 2009)

The function of the pyrolysis process is to produce a

maximum amount of condensable organic pyrolysis gas from
the biomass fed into the reactor. In the operation point
producing maximum oil yield the process is endothermic.
Typically 60 70% of the mass of biomass (dry base) is
converted to bio-oil, 10 20% to non-condensable gases and
10 20 % char residue. The heat value of the bio oil varies
from 13 to 18 MJ/kg and it consists of organic compounds
(app 90%) and pyrolysis water (app. 10%). The energy yield
of bio-oil is about 50 60% of the energy content of dry
biomass.

m oil

k=2

xi ,k
Tn ,k

xi ,in
Tn ,in

m fluidgas

m cw
xi ,k
Tn ,k

m tot

Tscr
mgas
moil

Tcw

k=3

2.1 Fast Pyrolysis

Condenser
model

k = element
i = component
n = phase

k=1

Fig. 2 The block diagram of the fast pyrolysis process model.

(Kaivosoja and Majanne, 2009)

8284

Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

Substances flowing in the pyrolysis process have been

divided into three phases: sand, biomass, and gas. These
phases are divided into ten components presented in a table 1.
Dashed arrow represents evaporation of moisture during
drying of biomass and solid arrows represent pyrolysis of
active material into pyrolysis products.
Table 1. Phases and components.

Flue gas

Temperature T
of phase n
in element k

Mass fraction x
of component i
in element k

Component i

Tsand,k

xsand,k

Sand

xash,k

Ash

xwater,k

Moisture (water)

xact,k

Active raw material

xchar,k

Char

xcond,k

Organic condensable pyrolysis gas

xnoncond,k

Non-condensable pyrolysis gas

xpyroH2O,k

Pyrolysis water (steam)

xsteam,k

Moisture (steam)

xfluidgas,k

Inert fluidization gas

Tbio,k

Tgas,k

and attemperators. In addition steam pressure, steam

temperature, and feed water control loops are included in the
model. The model is based on the mass, energy, and
momentum balances together with physical behaviour of heat
transfer, reaction kinetics etc. The structure of the CFB
furnace is represented in Fig. 3. In the model the furnace is
divided into nine blocks.

9
Cyclone

7
6
Solids
return
leg

Mass balances of the components xi of the active material mact

and pyrolysis products in the block k are:
(1)

The reaction rates of the pyrolysis reactions are defined by

the Arrhenius expression:
(2)

Calculation of experimental kinetic parameters A and E are

presented in Lehto (2007). Parameters yi define the
composition of the pyrolysis products. Values vary with
reaction temperature according to Bridgwater (1999). Values
of yi are between zero and one, and their sum must be exactly
one.
According to Milosavljevic (1996), the reaction enthalpy qpyr
of pyrolysis reactions can be expressed as a function of char
yield ychar:
q pyr 538 2000 ychar .

2.2 Circulating Fluidized Bed Boiler

A dynamic model of the CFB boiler consists of an air-flue
gas process consisting of air preheaters, furnace, and heat
exchangers and a water-steam process consisting of water
preheaters (economizers), drum, evaporator, superheaters,

Secondary air

Fig. 3. CFB boiler structure (Majanne and Kykk, 2009).

In the CFB boiler combustion takes place in a fluidized sand
bed circulating around the furnace hot loop. Bed sand and
fuel comprise a solid matter suspension. Combustion and
heat transfer efficiencies depend on the mass of the
circulating bed and the local suspension density in the
furnace.
The suspension density varies as a function of the height in
the furnace and the amount of the fluidizing air. The
fluidizing air flow is a nonlinear function of the boiler load.
Static suspension density profile as a function of the furnace
height h is described by an experimental equation

h f h ,

(4)

where (h) is a suspension density at height h, is a scaling

factor and f(h) is the fitting function for the density profile
(Breitholtz et al, 2001).

f h susp , x susp ,exit e Khtot e ahbelow susp ,exit e Khabove , (5)

(3)

More detailed description of the process and dynamic model

equations are presented in (Kaivosoja and Majanne,2009)

Primary air

the elements of the reactor. It can be calculated as a sum of

sand, biomass and fluidization gas feeds.

kv A e E RuT .

tot is supposed to be equal in all

Total modelled mass flow m

dxi ,k
m tot ,k xi ,k 1 xi ,k yi k v mact ,k .
dt

3
Fuel feed

mtot ,k

where susp,x is the suspension density at the bottom of the

furnace, susp,exit is the suspension density at the top of the
furnace, htot is furnace height, a and K adjusting parameters
for the density profile below and above point h, h below and
habove furnace heights below and above point h. Values for
susp,x, susp,exit, a, and K were determined by the process
experiments as a function of primary air mass flow.

8285

Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

During the transients caused by the change of the primary air

flow the dynamics of the change of the density profile is
determined as
d
1
h, t h, t f h, t ,
dt

Flue gase flow

(6)

O2 +
flue gases

Heat power
to wall and
radiation
superheaters

The combustion model is suitable for different types of fuels.

The quality of the fuel is parameterized by the elemental
composition of the fuel, moisture, proportion of volatile
matter, ash content, upper heat value of charcoal, and upper
heat value of dry fuel. The combustion model describes the
vaporization and combustion of volatile mater and
combustion of charcoal.
matter

is

mvol ,k kvol yvol ,k xONvol

2,k ,

modelled

by

O2 +
flue gases

Heat power
to wall

(7)

Volatiles,
charcoal

Element
temp.

AIR
PREHEATER

Secondary air

Cyclone temperature

Mixing
of sand

Sand flow

LOOP SEAL

Sand flow

ELEMENT 1

Primary air

Refractory
temperature

Recirculation
gas

Drum
pressure
Feed water flow
Feed water enthalpy

FEED WATER
PREHEATER

Heat power

Tube
temperature

(8)

Steam flow to primary

superheater

Tube
temperature
Feed water flow
Feed water enthalpy

Water-steam
mixture enthalpy

DRUM

Water-steam
mixture density

EVAPORATOR,
NONREFRACTED
WALL

Steam quality

Water-steam
mixture enthalpy

Steam quality

EVAPORATOR,
REFRACTED WALL
Refractory
temperature

Down comer flow

Drum water enthalpy

Fig. 5 The block diagram of the CFB boiler water side model.

For fine particles combustion speed tcf is

1,5 kc O 2

ELEMENTS 2 - 9

Heat power
to preheater

Tube
temperature

Sand
flow

Fig. 4 The block diagram of the CFB boiler flue gas side
model.

where c is a char density, dc is a particle diameter, Di is a

molecular diffusion coefficient from oxygen to nitrogen,
O2() is oxygen density far from the particle surface, and Sh
is Sherwood number.

fdf

Element
9 temp.

Refractory
and tube
temperature

Fuel (amount +
properties)

an

Combustion of charcoal is divided into combustion of big

and fine particles. Big particles are attritted into fine ones due
to the grinding effect between solid particles in the fluidized
bed. Combustion of big particles is constrained by diffusion
of oxygen and combustion of fine particles is constrained by
reaction speed (Borman, 1998). Combustion speed tc for big
particles is

tcf

Volatiles
charcoal

Heat power to conv.

superheaters and
economizer

FLUE GAS
DUCT

Sand temperature

where mvol,k is the amount of reactive volatile components in

element k, kvol is a reaction speed, yvol,k is a mass proportion
of volatile matter, xO2,k is a mole fraction of oxygen in
element k, and Nvol is an order of the reaction.

c d c2
tc
,
3Di O 2 Sh

Refractory
temperature

CYCLONE

Heat power
to wall

where is a time constant.

Combustion of volatile
experimental equation

Convective superheaters
and economizer tube
temperatures

Flue gase temp

(9)

where f is fine char density, df is fine char particle diameter,

kc is kinetic reaction speed factor determined by Eq. (2).
Solid matter and flue gases are separated from each other in
the cyclone. Hot flue gases are flown further to the
convective heat transfer section consisting of evaporators,
superheaters and preheaters.
Solid matter consisting of bed sand, ash and unburned fuel is
returned back to the bottom of the furnace via cyclone and a
loop seal. The function of the loop seal is to prevent counter
current gas flow through the return leg to the cyclone. Block
diagrams of the flue gas and water side models are shown in
figs. 4 and 5. More detailed description of the process and
the dynamic simulation is presented in (Majanne and
Kykk, 2009).

2.3 Interconnections of Engaged Processes

The pyrolysis and the boiler processes are connected together
by mass and energy balances. Thermal energy required by the
pyrolysis reaction is extracted from the boiler process as app.
800C hot sand taken from the return leg of the furnace hot
loop. In the pyrolysis reactor the required amount of hot sand
is mixed with biomass to maintain the optimal 500C
temperature in the reactor.
Sand and char are returned from the reactor to the boiler. This
return flow effects on the energy and mass balances of the
boiler. Returned sand is cooling the furnace because sand
temperature is app. 400C below the furnace temperature. On
the other hand the returned charcoal has a high energy
content, app. 32 MJ/kg, effecting on the released thermal
power in the furnace. Approximately 40 % of the energy
content of the pyrolyzed biomass is contained in the
remaining charcoal. It should also be taken into account that
in case of some disturbance in the pyrolysis process all the
biomass fed in the reactor might go unprocessed through the
reactor to the boiler.
After the reactor pyrolysis gases are condensed into a liquid
phase in the condenser. Some light hydrocarbon compounds
of the pyrolysis gas are non-condensable in the process

8286

Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

atmosphere, e.g. CO, CO2, C2H6(ethane) and C2H4(ethylene).

This non-condensable gas fraction is recirculated and used as
an oxygen free fluidizing gas for the reactor (Fig. 1).
However, those non-condensable gases are produced all the
time while running the process and the excessive amount of
gases is fed into the furnace of the boiler. This gas has
combustible components effecting on the energy balance of
the boiler. Gas flow effects also on the mass balance of the
flue gas flow of the boiler. Disturbances in the returned gas
flow effect on the convective heat transfer, furnace pressure
and further to the combustion air flow. Furnace pressure
measurements are typically connected with the boiler safety
automation and because of this connection, disturbances in
the furnace pressure may lead to boiler shut down.
3. BOILER CONTROL

dynamics of the heat transfer process from the furnace to the

evaporator.
In case of the pyrolyzer trip the heat power released by the
charcoal and non-condensable gases must be replaced by the
primary fuel feed as quickly as possible. By adding the feed
forward signals to the power control loop, speed of the
primary fuel feeders can be increased immediately after the
pyrolyzer fuel feed is stopped. Feed forward signals are
applied also in the steam temperature and furnace pressure
controls.
Fig. 6 shows open loop responses from the pyrolyzer trip to
the generated steam power in three cases, where the thermal
capacity of the pyrolyzer is 1/3, 2/3 and 1/1 of the thermal
power of the boiler. The controllable fuel feed to the boiler is
kept constant during the simulation runs.

The operation of the integrated pyrolysis process system must

be designed so that start ups, shut downs and disturbances in
the pyrolysis process will not cause major problems in the
operation of the boiler. The boiler process is anyway the main
process generating heat and power for the customers on-line.
Thermal power of the pyrolyzer must be scaled with the
thermal power of the boiler so that disturbances originated
from the pyrolyzer can be stabilized by the boiler control
system. Normal start ups and shut downs of the pyrolyzer can
be operated smoothly so that the boiler control system can
easily compensate these disturbances. The two worst cases
are the tripping of the pyrolyzer running full load and if the
biomass is blown unprocessed through the pyrolyzer into the
boiler.
The function of the combustion control is to feed a required
amount of fuel and combustion air into the boiler furnace.
The required fuel power is determined by the boiler master
controller according to the generated power or steam pressure
demand. The controllable part of the fuel power is fed into
the furnace by the fuel feeders and the rest is coming from
the pyrolysis process. The required combustion air flow is
calculated from the total fuel flow into the boiler and
trimmed by a flue gas oxygen content controller.
In order to stabilize the operation of the boiler measurements
from the pyrolysis process must be included to the boiler
control scheme. Operation state of the pyrolyzer must be
connected with the boiler fuel feed, combustion air, steam
temperature, and furnace pressure controls. The easiest way
to include the interactions between the pyrolyzer and the
boiler is to apply filtered feed forward signals from the
biomass feed of the pyrolyzer and non-condensable gas flow
to the boiler controls. The ideal transfer function Gff_ideal for
the feed forward control is

G ff _ ideal

Gd
,
Gp

Steam power

100

1/1
2/3
1/3
90

80

70

60

50

60

70

80

90

100

110

120

130

140

150

160

Time, min

Fig. 6. Effect of the pyrolyzer thermal capacity compared

with the boiler capacity to the generated steam power during
the pyrolyzer trip.
Fig. 7 shows the closed loop responses from the pyrolyzer
trip to the generated steam power with the same three
different size pyrolyzers. Combustion control loop with feed
forward signals from the biomass feed to the pyrolyzer and
non-condensable gas flow to the furnace stabilizes generated
disturbances quite satisfactorily.
Steam power

100

1/1
2/3
1/3
99

98

(10)
97

where Gd is the disturbance transfer function from the fuel

power coming from the pyrolyzer to the boiler power and Gp
is the transfer function from the boiler fuel feed to the boiler
power. In the following simulations only static feed forward
transfer function is applied because the dynamics of the fuel
flows from the pyrolyzer and from the fuel feeding system to
the boiler are very similar and fast compared with the

96

50

55

60

65

70

75

80

85

90

95

100

Time, min

Fig. 7. Effect of the pyrolyzer size to controlled steam power

during the pyrolyzer trip.

8287

Preprints of the 18th IFAC World Congress

Milano (Italy) August 28 - September 2, 2011

Fig. 8 shows the performance of the feed forward

compensation in the steam pressure control during the trip of
the pyrolyzer. The thermal capacity of the pyrolyzer is 1/3 of
the thermal power of the boiler. The first case depicts the
control response with no feed forward compensation. In the
second case the feed forward information is connected to the
controller and the maximum rate of the change of the fuel
feed is limited to 10%/min. In the third case the rate of the
change of the fuel feed is unlimited.
Live steam pressure

+3

Feed forward control with limited rate of change

Feed forward control with unlimited rate of change
Control with no feed forward

+2

+1

% -1

-2

-3

-4

-5

50

55

60

65

70

75

Time, min

80

85

90

95

100

Fig. 8. Performance of the power control loop of the boiler

without feed forward information (- - - -), with feed forward
and limited rate of change of fuel feed (____), and with feed
forward and unlimited rate of change of fuel feed (----).
4. CONCLUSIONS
Fast pyrolysis is a novel process for converting solid bio
mass to liquid form to replace heavy fuel oil or as a raw
material for refinery processes to produce higher value bio
fuels like bio diesel. Dynamic simulation has been used to
help designers to evaluate different options in process
structures and control principles of this new process. A very
important point of view is how this new process interacts
with the interconnected power boiler. The primary function
of the power boiler is, however, power generation, not to run
pyrolysis process. The important issues to be considered are
the disturbance propagation from the pyrolyzer to the boiler
and the requirements for the operation range of the boiler to
produce enough hot sand for the pyrolysis process. Optimal
scaling of the power ratio between the pyrolyzer and the
boiler is required to result stable operation of the boiler and
maximum yield of the bio-oil.
This paper presented some results gained in the research
project with Metso Power and Tampere University of
Technology. This project is a part of the long term research
work dealing with dynamic modelling of different type of
fluidized bed boilers and new processes for production of bio
energy.

Bridgwater, A.V. (1999). Principles and practice of biomass

fast pyrolysis processes for liquids. Journal of Analytical
and Applied Pyrolysis, 51(12):322.
Bridgwater, A.V. and Peacocke G.V.C. (2000). Fast pyrolysis
processes for biomass. Renewable and Sustainable
Energy Reviews, 4(1):173.
Calabria, R., Chiarello, F., De Bellis, V., Massoli, P. (2005).
Combustion Fundamentals of Pyrolysis Oil Based Fuels,
Instituto Motori CNR, Napoli, Italy.
Galgano, A., Salatino, P., Crescitelli, S., Scala, F. &
Maffettone, P.L. (2005). A model of the dynamics of a
fluidized bed combustor burning biomass. Combustion
and flame Vol.140, No.4, pp. 271284.
Kaivosoja J. (2008) Modeling and Control of Pyrolysis
Process Integrated with Circulating Fluidized Bed
Boiler. Master of Science Thesis (in Finnish), Tampere
University of Technology, Institute of Automation and
Control. Tampere 2008
Kaivosoja J and Majanne Y. (2009) Dynamic Modeling and
Simulation of Fast Pyrolysis Process Integrated with
Circulating Fluidized Bed Boiler. Proceedings of IFAC
Symposium on Power Plants and Power Systems
Control, 5-8.7.2009, Tampere, Finland
Kykk P. (2007) Dynamic modeling of the circulating
fluidized bed boiler. Master of Science Thesis (in
Finnish), Tampere University of Technology, Institute of
Automation and Control. Tampere 2007
Lehto, J. (2007). Development and Characterization of Test
Reactor with Results of Its Application to Pyrolysis
Kinetics of Peat and Biomass Fuels, Dissertation,
Tampere University of Technology, Publication 665.
Majanne, Y. and Kykk, P. (2009) Dynamic model of the
circulating fluidized bed boiler. Proceedings of IFAC
Symposium on Power Plants and Power Systems
Control, 5-8.7.2009, Tampere, Finland.
Milosavljevic, I., Oja, V., Suuberg, E.M. (1996). Thermal
effects in Cellulose Pyrolysis: Relationship to Char
Formation Processes, Industrial & Engineering
Chemistry Research, 35(1996), 653-662.
Saastamoinen, J.J. (2006). Simplified model for calculation
of devolatilization in fluidized beds, Fuel 85, 23882395.
Scala, F. & Salatino, P. (2002). Modelling fluidized bed
combustion of high-volatile solid fuels. Chemical
engineering science, Vol. 57, pp. 1175-1196.
Sipil, E., Vasara, P., Solantausta, Y., Sipil, K. (2007).
Feasibility and market potential of pyrolysis oils in
European pulp and paper industry, Proceedings of the
15th European Bioenergy Conference and Exhibition,
Berlin.

5. REFERENCES
Borman, G.L. & Ragland, K,W. (1998). Combustion
engineering. Singabore 1998, McGraw-Hill Inc. 613 p.
Breitholtz, A., Leckner, B. & Baskakov, A.P. (2001). Wall
average heat transfer in CFB boilers. Powder
Technology 120(2001), pp. 41-48.

8288