Accepted Manuscript

Cross-flowing displacement ventilation system for conveyor belts in

the food industry

Tobias Heidrich , Aria Alimi , Leon Grothues , Jens Hesselbach ,

Olaf Wünsch

PII: S0378-7788(18)30853-3
DOI: https://doi.org/10.1016/j.enbuild.2018.09.021
Reference: ENB 8807

To appear in: Energy & Buildings

Received date: 15 March 2018

Revised date: 5 June 2018
Accepted date: 15 September 2018

Please cite this article as: Tobias Heidrich , Aria Alimi , Leon Grothues , Jens Hesselbach ,
Olaf Wünsch , Cross-flowing displacement ventilation system for conveyor belts in the food indus-
try, Energy & Buildings (2018), doi: https://doi.org/10.1016/j.enbuild.2018.09.021

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and
all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Cross-flowing displacement ventilation system for

conveyor belts in the food industry
Tobias Heidrich a, Aria Alimi b, Leon Grothues c, Jens Hesselbach a, Olaf Wünsch b
a
Department for Sustainable Products and Processes (upp), University of Kassel, Kassel, 34125, Germany
b
Chair for Fluid Mechanics, University of Kassel, Kassel, 34125, Germany
c
Research and Development, Laboratory, Miele & Cie. KG, Gütersloh, 33332, Germany (formerly Department for
Sustainable Products and Processes (upp), University of Kassel, Germany)
heidrich@upp-kassel.de

T
Keywords: Local air conditioning, displacement ventilation in the food industry, computational fluid

IP
dynamics for air conditioning systems.

CR
ABSTRACT

In the following paper a local air conditioning concept for conveyor belts in the packaging sector of
chocolate production is described, comprehensively analyzed and evaluated with respect to its

US
suitability for a use case. For this purpose, extensive laboratory investigations were carried out and
energy savings and numerical flow simulations were calculated. Often the entire hall in the packaging
area of chocolate production must be strictly air-conditioned to fixed limits to avoid the appearance of
AN
condensation effects or melting of the product. Using the approach of conditioning only where it is
necessary, high energy saving potentials are achievable since temperature and humidity limits in the
hall can be significantly increased.
M

1. Introduction and problem description

ED

In the field of building air conditioning, high energy saving potentials can be raised in many ways [1-
3]. In addition to control engineering improvements and the further development of the various
components of an air conditioning system, also different air duct concepts offer considerable saving
potentials [4,5]. Above all, packaging halls of temperature-sensitive products are ideal for analyzing
PT

and implementing novel air guidance concepts, as the packaging process usually takes place on open
conveyor belts.
In the application case chocolate bars from the cooling passage reaches the conveyor belt in the
CE

packaging area at a temperature of around 14 to 15 °C and thus heats up slowly. Since deformations
of chocolate already occur at 18 °C and to avoid condensation effects on the product and quality
problems with the packaging cartonnage the entire packaging area must be air-conditioned to 18 °C
AC

and 50% relative humidity [4]. The strict temperature and humidity specifications lead to high energy
consumption in the hall air conditioning. The approach of local air conditioning pursues the goal of air
conditioning only an area, where it is really needed and thus makes it possible to increase the
flexibility of hall climate control [6,7]. Therefore, air-conditioning variants are being developed,
which enable air conditioning around the product-carrying conveyor belt in the packaging area. For
the rest of the hall, the fixed air conditioning limits can thus be significantly extended and based, for
example, on the limits of the comfort field according to Fanger and others. The scientific research of
Wagner [4,8], who investigated a full encapsulation around the chocolate bar carrying conveyor belt,
yielded high energy savings in the area of hall air conditioning. Another approach to local air
conditioning in the form of cross-flowing displacement ventilation, which flows laterally on the
conveyor belt, allows free access to the product at all times and significantly reduces the area to be
 ACCEPTED MANUSCRIPT

air-conditioned. In the following, it will be shown how the construction of a cross-flowing

displacement ventilation for conveyor belts can be realized on a laboratory scale. Furthermore,
computational fluid dynamic (CFD) simulation calculations are presented and compared with real
measurements, and energy saving potentials are calculated.

2. State of Research

The general approach of so-called local ventilation is applied in many ways. Olander et al. [9]
differentiates between fixed, flexible and mobile local ventilation modes. Fixed systems cannot be
changed except for possible opening and closing of enclosures. A fixed, local ventilation system can
be a local permanently installed conveyor belt air conditioning system or an extractor hood in the

T
laboratory. Flexible systems have movable manifolds or flexible hoses connected to an air

IP
conditioning system. An example of this is a mobile suction hose for the local removal of foreign
substances. Mobile systems can be placed at any suitable location. The air conditioning system for

CR
supplying conditioned air or removing contaminated air is then mounted on wheels or a portable
frame. Local conveyor belt air conditioning can also be designed as a mobile unit. [9]
In addition to the mentioned research work by Wagner, who investigated a fixed full encapsulation of
the conveyor belt for temperature-sensitive products, there are many other approaches to local air

US
conditioning of conveyor belts. Burfoot et al. [10] describes uni-directional air flow designs, open
troughs, semi-closed tunnels and fully-closed tunnels for local conveyor belt air conditioning in the
food industry.
AN
Uni-directional air flow designs have been used for several years in various industries such as the
pharmaceutical, electronics and food industries. For conveyor belt air conditioning in the food
industry, uni-directional supply air variants are designed so that the supply air flows from a diffuser or
textile hose above the conveyor belt in the direction of the product to be air-conditioned. In most
M

cases, the supply air flow is directed to the product via lateral boundaries on the supply air duct,
which extend to the head height of the persons working on the conveyor belt below. Disadvantages
ED

are that cold supply air hits the people working on the conveyor belt, relatively high supply air
volume flows are required, and the supply air volume flow is susceptible to transverse inflows from
the surroundings of the hall. Simple conveyor belt cleaning or free product access are clear
advantages in contrast.
PT

Open troughs for conveyor belt air conditioning consist of a supply air duct installed on one or both
sides of the conveyor belt. The product is air-conditioned via cross-flowing air. The cross-flow
displacement ventilation investigated in this publication belongs to this type of construction. A
CE

suitable discharge height and position of the supply air ducts and carefully selected supply air volume
flows are essential for energy-efficient operation and to ensure product safety. Relatively low supply
air volume flows and free product access are clear advantages of this variant. However, the effort
AC

required to clean the conveyor belt is comparatively high.

Furthermore, small tunnel systems can be installed around the conveyor belt for local air conditioning.
Such encapsulations can completely enclose the conveyor belt, as Wagner, among others, has
examined. Product access can be via flaps. It is also possible to partially encapsulate the conveyor belt
to ensure free product access from at least one side. CFD simulations and fog tests by Burfoot et al.
[10] and Heidrich et al. [7] show that conveyor belt encapsulation can completely prevent the entry of
foreign substances and transverse inflows. The product safety is correspondingly very high, but so is
the installation effort. Product access and conveyor belt cleaning are also comparatively difficult. The
required supply air volume flows and speeds are below the uni-directional supply air variant and
above open trough designs.
 ACCEPTED MANUSCRIPT

However, the degree of penetration of the described local conveyor belt air conditioning systems in
operational practice, especially open troughs in the food industry, is still in a preliminary stage. [10-
12]

In addition to product air conditioning, the approach of local air conditioning can also be applied to
people, then called personalized ventilation. For example, Ahmed et al. [13] investigated a new type
of local ventilation system for office buildings, whereby workstations are air-conditioned locally from
above. Assaad et al. [14] also investigated the local air conditioning of workplaces by means of a
frontal air supply. CFD simulations were used to investigate air flows and temperature distributions in
both cases and energy savings of up to 21.3% (ventilation from above) and up to 30% (frontal
ventilation) could be calculated.

T
IP
3. Laboratory construction and results

CR
To test the functionality as well as to define the optimal parameter sets for the operation of the cross-
flowing displacement ventilation for the packing sector of chocolate, extensive laboratory
investigations were performed. The construction is installed in a climatic chamber and is supplied
with conditioned air via a full-climate system.

US
An air outlet element of the displacement ventilation consists of several components, which are
described in the following figure.
AN
Flow meter with throttle
Air distributor nozzle
Perforated sheet
M

Filter mat
Perforated sheet
ED
PT

Bypass flap
Figure 1: Schematic construction of the displacement ventilation
CE

As can be seen, the air outlet element consists of a variably usable perforated plate, a variably usable
filter mat, an air distribution nozzle, a bypass-flap for fine adjustment of the inlet volume flow and a
AC

differential pressure volume sensor. The entire laboratory construction is illustrated in the following
figure.
 ACCEPTED MANUSCRIPT

Device for a differential pressure Supply air Manually adjustable slide for variable
volume flow measurement adjustment of the outlet height

T
IP
CR
Manually controllable inlet flap for finer Interchangeable perforated plate,
adjustment of the supply air volume behind interchangeable air filter
Figure 2: Laboratory construction of the cross-flowing displacement ventilation

US
Furthermore, the air outlet elements have been supplemented with manually adjustable slides to adjust
the outlet height. The supply air temperature is measured by two sensors at the lower edge and in the
AN
middle of the supply air outlet. Two more sensors are mounted in the middle of the conveyor belt. The
average value of all four sensors gives the local conveyor belt temperature. The constant general
conditions for the experimental investigation of the displacement ventilation are summarized in the
following table.
M

Table 1: Constant parameters in the experimental investigation of the displacement ventilation

Parameter Fixed value
ED

Hall temperature 21 °C (± 1 K)
Hall air humidity 50% rel.
PT

Upper temperature limit on the product 18 °C

Supply air humidity 50% rel.
CE

Filter class in the outlet element F7

Hole diameter of the inserted perforated plate as air outlet 8 mm
AC

The supply air temperature and the supply air speed are varied at different outlet heights. The ratio
1:4, 1:6 or 1:8, which is used in this paper, means that the conveyor belt is four times, six times or
eight times longer than the outlet height. This ratio considerably simplifies the transferability to a later
practice. In addition, experiments were performed in which an increased humidity in the hall ambient
air was brought in via spray fog. To obtain a certain air outlet velocity, the supply air volume flow as
well as the outlet area of the displacement air outlet must be adapted. The outlet area is the product of
the length (here 2 m) and height of the displacement air outlet. The height can be variably changed by
the push-device at the outlet itself and adjusted in a corresponding relation to the width of the
conveyor belts. The width of the band is 0.86 m. The volume flow can first be roughly controlled by
the air conditioning system and further fine-tuned by the throttle- and bypass-flaps installed at the
outlets, as initially described. For this purpose, the volume flows are manually measured and
 ACCEPTED MANUSCRIPT

compared. The specific supply air volume flows (in m³/h) at defined outlet velocities over the
conveyor belt are shown in the following table.

Table 2: Specific supply air volume flow at a defined discharge speed and the ratio of the outlet
height to the width of the conveyor

Ratio H:B
1:4 1:6 1:8
Outlet
(Height 21,5 cm) (Height 14,3 cm) (Height 10,8 cm)
velocity

0,025 m/s 39 m³/h 26 m³/h 19 m³/h

T
0,05 m/s 77 m³/h 51 m³/h 39 m³/h

IP
0,1 m/s 155 m³/h 103 m³/h 77 m³/h

CR
0,2 m/s 310 m³/h 206 m³/h 155 m³/h

The results of the test series with an outlet ratio of 1:8 are visualized in the following figure. Shown

US
are the resulting average conveyor belt temperatures at different supply air temperatures and outlet
velocities at a constant outlet ratio of 1: 8.
AN
Conveyor belt temperature in ° C

19

18
0,025
ED

17
0,05
16
0,1 Outflow velocities
15
PT

in m/s
14 0,2
18 17 16 15 14
CE

14-15 15-16 16-17 17-18 18-19

Supply air temperatures in ° C

Figure 3: Results displacement ventilation concept – conveyor belt temperatures in dependence of the
AC

supply air temperature and different air outlet velocities at an outlet height-to-transport bandwidth-
ratio of 1: 8

If the deviation/spread of the real temperature course (supply air temperature to conveyor belt
temperature) from the ideal line (supply air temperature corresponds to conveyor belt temperature) is
considered closely at different outlet velocities, then subsequent results can be shown (here selection
for outlet ratio 1:6).
 ACCEPTED MANUSCRIPT

22
21
Conveyer belt temperature in °C

20
19
18
temperature curve - velocity 0.025 m/s
17 R² = 0.9985
ideal temperature curve
16
temperature curve - velocity 0.1 m/s
15 R² = 0.9991
trend line - velocity 0.025 m/s
14
trend line - velocity 0.1 m/s

T
13
13 14 15 16 17 18 19 20 21 22

IP
Supply air temperature in °C

CR
Figure 4: Conveyor belt temperatures under varying supply air temperatures at an outlet velocity of
0.025 m/s and 0.1 m/s with an outlet ratio of 1:6

US
Using the determination factor R² as an example, it can be shown that the average conveyor belt
temperature is almost linear to the supply air temperature. Furthermore, a higher supply air velocity
clearly leads to a smaller spread between the real and the ideal temperature course. Thus, at a supply
air temperature of 14 °C and an increase of the outlet velocity the temperature difference can be
AN
reduced from 0.025 m/s to 0.1 m/s by about 0.6 K.
When considering the different outlet ratios, a high outlet height leads to slightly better results than a
lower one. Due to the increase of the outlet area, the volume flow increases by a constant outlet
M

velocity. As a result, the ambient air is being displaced more strongly and thus the effects from the
environment are being reduced. Overall, the influence of the outlet height can be maintained, but it
has a less pronounced effect on the local temperature structure compared to the variation of the supply
ED

air temperature and velocity.

Furthermore, fog tests were performed to visualize air flows. In the first case, the fog was guided from
the external environment on the conveyor belt, in the second case the fog was blown internally from
PT

the air outlets on the conveyor belt. Supply air volume flows and outlet ratios are varied several times
in these experiments. A selection of the results can be seen in the following pictures.
CE
AC
 ACCEPTED MANUSCRIPT

Outlet velocity: 0.2 m/s, ratio: 1:8

Fast fog displacement

Relative
stabile layering

Outlet velocity: 0.025 m/s, ratio: 1:8

T
IP
Increased fog

CR
entry in the edge
region
Figure 5: Fog test – Fog external

US
As shown, at a high supply air velocity (0.2 m/s) with an external contribution of fog, only a small
amount of fog enters the displacement ventilation area. Overall, a relatively stable stratification can be
seen at the outlet height. A slight entry takes place at the edge area of the displacement ventilation
AN
outlets. This entry over the edge area increases with decreasing supply air velocity (to a minimum of
0.025 m/s). At low supply air velocities, the fog also remains longer over the conveyor belt area. The
reason why the entry increases at low supply air velocities is made clear by the second experiment, in
which the fog is injected directly (internal) into the right-hand displacement air outlet. At a high
M

supply air velocity (0.2 m/s), the flow of cool supply air is much more targeted. The outlet pulse of
the air is so great that the flow reaches the middle conveyor belt area where it meets the supply air of
ED

the opposite displacement air outlet and thus rises in the middle.

Outlet velocity: 0.2 m/s, ratio: 1:8

PT

High impulse and

CE

purposeful outflow
AC

Outlet velocity: 0.025 m/s, ratio: 1:8

Low impulse outflow

Figure 6: Fog test – Fog internal

At low outlet velocities, the pulse is lower. The supply air gets increasingly mixed with the ambient
air, as can be seen well in the previous figure.
 ACCEPTED MANUSCRIPT

In a further step, the contribution of water in the form of a spray fog into the displacement air area
was simulated. The water was finely sprayed over the conveyor belt at different volume flows and
brought into the air. The aim was to check how far the fog penetrates the displacement air area or the
water droplets condense on the conveyor belt. Thereby conclusions about other things can be drawn,
such as hygienic aspects (for example, emulation of an employee on the conveyor belt). The recording
was carried out in this experiment visually and by checking humidity/droplets on the bottom of the
conveyor belt. The experiment was carried out with the variation between the minimum volume flow
(outlet ratio 1:8, outlet velocity 0.025 m/s) and the maximum volume flow (outlet ratio 1:4, outlet
velocity 0.2 m/s). The results of the experiment series are clear. All tested system settings - from a
low exit velocity at a low outlet height to a high exit velocity at maximum outlet height - show little
or no change in relative humidity. The cross-flowing displacement ventilation is therefore an optimal

T
system for keeping humidity limits locally. The investigation of the fogging experiments regarding

IP
the infiltration of water droplets also showed clear results. Neither very small nor very large volume
flows can prevent the penetration of water droplets into the displacement air area. The gravity of the

CR
droplets is greater than the upward directed impulse force of the ascending airflow. The displacement
ventilation is therefore not suitable for avoiding the infiltration of water droplets in the air-conditioned
area. Only an active suction of foreign particles can be avoided by the constant outflow of air from the
displacement air area.

US
Depending on the supply air volume flow and velocity, the displacement ventilation system for local
product air conditioning provides good to very good results in terms of keeping the temperature
AN
limits. Overall, the average temperature increase on the conveyor belt is usually less than 0.5 K
compared to the outlet temperature of the displacement ventilation. From a supply air velocity of 0.1
m/s, a very stable temperature level can be seen with a small difference to the supply air temperature.
Lower supply air velocities (0.05 m/s - 0.025 m/s) also lead to positive results with deduction of
M

collateral. In summary, regarding sufficient product safety, a stable temperature level and desired
energy efficiency, an outlet ratio of 1:8, an inlet air velocity of 0.1 m/s and an inlet air temperature of
ED

16 °C crystallize out for optimum adjustment of the displacement ventilation parameters.

4. Fluid-mechanical simulation
PT

CFD simulations are already widely used in the food industry. CFD simulations offer many
advantages for understanding the basic physical and fluid-mechanical processes, for example in the
areas of drying, sterilization, mixing and cooling and similar processes. Product design can also be
CE

improved based on the results of CFD simulations. [15]

CFD simulation to investigate the flow and temperature behaviour of the air in the ventilation supply
of buildings [16-19] and of conveyor belts to be air-conditioned [10] is an established method.
AC

Consequently, CFD simulations are carried out to investigate the fluid mechanical behaviour of the
air, the heating behaviour of the chocolate bars and to validate the data measured in the laboratory
tests.

4.1 Model description

The calculations have been performed using ANSYS FLUENT 17.1 and the results are post processed
in ANSYS CFD Post. The models have been meshed using GAMBIT meshing software. The model
consists of two phases, solid phase, which are the chocolate bars and the fluid phase being the air. For
the fluid phase, the flow must be calculated alongside with the temperature. For the solid part though,
the conservation of energy would be solved.
 ACCEPTED MANUSCRIPT

To simulate the flow numerically, Navier-Stokes equations would be solved using Finite-Volume-
Method. On the other hand, calculating the temperature for both solid and fluid phase requires solving
the equation of energy numerically. Continuity Equation (1) and the Equation of Motion (2) are
shown below:

( ) (1) Continuity Equation

( ) (2) Equation of Motion

Where ρ is the density of air, u is the velocity vector of the flow, p is the air pressure, μ is the dynamic
viscosity and f is external forces per unit mass. Regarding the temperature, equation (3) shows

T
conservation of energy:

IP
( ) ( ) (3) Conservation of Energy

CR
Where ε is the thermodynamic internal energy, T is the temperature and KH is the heat conduction
coefficient. These equations are calculated numerically for a model shown in figure 7. The room has

US
been modeled using ANSYS Design Modeler. It represents the dimensions and distances used for the
experiments described before.
AN
M
ED
PT

Figure 7: Room geometry used for numerical simulations

According to figure 2, there can be two symmetry planes assigned to the room used for the
CE

experiments: one between the chocolate bars and the other one cuts the chocolate bars in half. Hence,
to reduce the calculation time, a quarter part of the room shown in figure 7 is modeled and meshed.
The blue walls show the symmetry planes assigned to the model. The red wall represents the air outlet
AC

element, from which the cold air flows onto the conveyor belt. The brown walls are the interfaces
between the chocolate bars and the room. The green surface represents the conveyor belt where the
chocolate bars are located. As the chocolate pieces sit on the conveyor belt close to each other, the
model includes multiple long solid parts on the conveyor belt representing the chocolate pieces. There
are four chocolate bars in the middle of the room. Considering the fact that there is no top roof closing
the conveyor belt area, almost the whole hall has been modeled in a large dimension and all other
areas are assigned as areas where the air can escape (shown in white). The whole geometry has
approximately 164000 orthogonal cells. For the simulations, air is assumed to behave as an ideal gas.
In this way, the effect of natural convection can be also considered in the calculations. Regarding the
solid part, the material properties of the chocolate are listed in table 3 according to Wagner [4]:
 ACCEPTED MANUSCRIPT

Table 3: Chocolate properties

Constant Value
Density 1270
Cp (Isobaric Specific Heat Capacity) 3140 ( )
Thermal Conductivity ( )

4.2 Initial and boundary conditions

For the boundary conditions, the surrounding walls have a backflow temperature of 21 °C. The

T
bottom surface of the conveyor belt is assumed to be adiabatic with an initial temperature of 21 °C.
Two blue sides are bounded with symmetry plane as already discussed. The room has also an initial

IP
temperature of 21 °C. The chocolate bars have a temperature of 15 °C when they enter the packaging
area. The air flows onto the conveyor belt with constant velocity of 0.1 m/s from the air outlets like

CR
the experiments with a temperature of 16 °C. The free surface of the chocolate is coupled with the air
and the heat energy can be transferred from air to chocolate and vice versa. Due to a turbulent
behavior of air flow especially near the air outlet elements, k-ω standard model is used for calculation
of the flow.

4.3 Results US
AN
The calculation results are achieved for the steady-state and the results are shown and discussed in the
upcoming figures.
M
ED
PT
CE

Figure 8: Velocity field

AC

Figure 8 shows the velocity field at the symmetry plane for the steady state solution. The black lines
show the stream lines, which initiate from the air outlet elements. They show that the air flows
horizontally from left and right air outlets toward the middle of the room and starts to arise toward the
top area where the hall roof is located. As the air flows from both outlets reach each other, small
vortices will be developed. These could be seen in detail in figure 12. Furthermore, the air flow from
the outlets has almost no influence on the regions far from the conveyor belt in right and left side of
the hall. To study the temperature regime onto the conveyor belt, the steady-state temperature field is
shown in figure 9.
 ACCEPTED MANUSCRIPT

T
Figure 9: Results – Temperature field - CFD-Simulations – Local air conditioning via displacement

IP
ventilation at a hall ambient temperature of 21 °C, a supply air temperature of 16 °C and a supply air
velocity of 0.1 m/s

CR
At a supply air velocity of 0.1 m/s with a supply air temperature of 16 °C, a chocolate bar temperature
of 15 °C, as it occurs after the cooling corridor in the packaging hall, and a hall ambient temperature

US
of 21 °C, the chocolate can be sufficiently tempered. These results coincide with the laboratory tests.
AN
M
ED
PT

Figure 10: Contour of temperature in chocolate bars

CE

Figure 10 shows the temperature contour on the symmetry plane of two chocolate bars shown in
figure 7. The maximal temperature is on the sides near the air outlet elements. The reason can be
explained using the room temperature contour shown in figure 9. There can be noticed that the area
AC

near the air outlets is still influenced by the 16 °C outlet air flow which is higher than the initial
temperature of chocolate bars. To check the thermal heating behavior of the chocolate bars a time
dependent plot of temperature inside the chocolate bar were carried out. This is important, for
example, if the local air-conditioning fails and the bar is on the conveyor belt for a short time in the
hall environment (assumed to be 21 °C in this case). For this reason, a transient calculation is
performed, and the temperature trend is shown in figure 11.
 ACCEPTED MANUSCRIPT

T
IP
CR
Figure 11: Chocolate temperature trend

The red line is the volume averaged temperature over the whole volume of bars. The blue line is the

US
area averaged temperature trend on the symmetry plane of chocolates. They both seem pretty
satisfying and are under the melting temperature of the chocolate. Besides, the evolution of
temperature has a very slow velocity. It shows a 0.05 °C variation in about 12 minutes. It is more than
AN
the time needed for processing the chocolate bars as they pass though the hall on the conveyor belt.

Comparing the flow profile of the previously described fog experiments with the flow profile of the
M

CFD simulations, the following can be shown for the parameter combination outlet velocity 0.025
m/s, outlet temperature 16 °C and outlet ratio 1:8.
ED

Ambient air is warmer and not stationary

PT

Increased turbulence and air entry in the edge area

CE

Flow field less homogeneous

Variances at
the air outlet
AC

Heat input through conduction and convection via the conveyor belt

Figure 12: CFD- flow profile of the air compared to the laboratory results

The figure shows, that at a supply air speed of 0.1 m/s and for chocolate bars, which are not moving in
the middle of the conveyor belt, small air vortexes occur. However, these vortexes, as previous
simulation and measurement results show, do not affect the air-conditioning of the chocolate bars.
It can also be seen that differences exist between the results of the experimental investigations and the
CFD simulation. Overall, the flow field and the temperature profile in the practical implementation of
 ACCEPTED MANUSCRIPT

the displacement ventilation are less homogeneously pronounced. This is justified by the increased
displacement air vortex in the air and increased influences from the environment. While these
influences increase relatively at low supply air velocities, the influence at higher volume flows (≥ 0.1
m/s) is less significant. However, the basic processes and flow courses of simulation and laboratory
tests are congruent.

5. Energetic simulated calculation for an application case

To calculate the energy requirement of the displacement ventilation and the corresponding savings for
the packaging area of the chocolate bar production in the application case, extensive models of the
packaging hall of the food company, the air conditioning system, the cooling supply and the steam

T
supply, which is used for humidification, were created in MATLAB/SIMULINK.

IP
The picture below shows a scheme of the packaging sector of the confectionery manufacturer.

Packaging area

CR
HVAC

Local climatization
Production hall

29 °C
Conveyor belt
US
AN
Cooling pathway Corridor
25 °C
M

Figure 13: Scheme of the packaging sector of the confectionery manufacturer (based on Schirmer et
ED

al. 2016)

The building model consists of coupled, ordinary differential equations and contains the real wall
PT

structure of the hall. Internal load dependent on the production plan is determined by real active
power measurements of the machines in the hall, calculations for the heat of illumination according to
VDI 2078 [20], humidity and heat emission of the persons in the hall according to VDI 2078 [20] and
CE

the influence of the local air conditioning by transfer of the laboratory results. The room to be air-
conditioned has an area of approximately 2000 m² with a height of 8.5 m. The room ventilation plant
model is based on the physical laws of the Mollier diagram, for the cooling supply (compression
AC

chiller with flow temperature of -1 °C) and steam humidification (3 bar of saturated steam)
characteristic curves of the manufacturer are used. All the used simulation models were validated by
real measurements on a laboratory scale or by measurements in the confectionary company
[4,6,8,21,22].
The annual energy demand for climate control of the packaging area without local air conditioning
and rigid temperature and humidity limits is shown in Table 3 and based on the data from the installed
energy monitoring system.

Table 4: Energy requirement of the scenario Status quo for a packaging hall in the confectionary
manufacturing
 ACCEPTED MANUSCRIPT

Scenario El. power in kWhel/a Heat in kWhth/a Cold in kWhth/a

Status quo 1.024.765 568.884 453.552

Due to the local air conditioning, the temperature and humidity limits in the hall can be extended in
the simulation from a rigid 18 °C and 50% relative humidity to a variable 18 °C to 21 °C and 30% to
65% relative humidity. The weather data used for the simulations come from the year 2016 for a
central German location. To clearly describe the energy-saving potential of a cross-flowing
displacement ventilation system, many simulation calculations with different parameters were carried
out. The outlet height of the displacement air diffuser (outlet ratio 1:4, 1:6, 1:8), the supply air

T
velocities (0.02 m/s, 0.05 m/s, 0.1 m/s, 0.2 m/s) and the supply air temperature (14°C to 17°C in 1

IP
Kelvin steps) were varied. The following table provides an overview of the maximum and minimum
savings that can be achieved as a result of the calculations.

CR
Table 5: Maximum and minimal achieving savings through the installation of the source ventilation
Displacement air
parameter
Saving
MAX /
MIN
Outlet ratio;
Velocity;
Temperature
US Saving
el. Energy
Saving
Heat
Saving
Cold
AN
El. MAX 1:6; 0,2 m/s; 14 °C 16% 28% 28%
Energy MIN 1:8, 0,025 m/s; 17 °C 15% 48% 50%
MAX 1:6; 0,025 m/s; 16 °C 15% 48% 50%
Heat
M

MIN 1:4; 0,2 m/s; 14 °C 15% -3% 14%

MAX 1:8; 0,025 m/s; 17 °C 15% 48% 51%
Cold
MIN 1:4, 0,2 m/s; 14 °C 15% -3% 14%
ED

For the favored parameter combination (supply air temperature: 16 °C, outlet velocity: 0.1 m/s, outlet
ratio 1:8), electric power savings of 15%, heat savings of 45% and cold savings of 46% can be
PT

reached.

As Table 5 also shows, the highest variance of the calculated savings is in the area of heat
CE

consumption. Overall, thermal energy savings increase with decreasing supply air volume flows and
rising supply air temperatures. In contrast, high volume flows and low supply air temperatures result
in less significant savings. This initially has a contradictory effect since the heat requirement should
AC

theoretically be lower at low supply air temperatures. However, since the cold supply air of the
displacement ventilation flows into the production hall, it must be tempered to at least 18 °C, which
leads to increased energy consumption. Both effects (savings in displacement ventilation and
increased heating requirements of the hall) thus cancel each other out. In addition, there is the added
conditioning effort of displacement ventilation (specifically reheating of the supply air after
dehumidification), which increases with the level of the volume flows.
The need for cold is also subject to a certain variance. Like the heat requirement, the highest savings
are achieved at low volume flows and high supply air temperatures. Especially in the warm summer
months, there is a clear reduction in cooling requirements. This is not only influenced by the reduced
cooling requirement of the hall, but also by the reduced requirement for dehumidification of the hall
air.
 ACCEPTED MANUSCRIPT

The savings in electrical energy are mainly due to the reduced need for cold and thus the lower
utilization of the compression refrigeration machine. The percentage savings amount to around 15 %
of total electrical requirements. It is striking that the electrical savings are not significantly influenced
even with parameter settings that lead to low cold savings. This can be explained by the fact that an
advantageous operating point of the chillers is already achieved at low cold savings.

Burfoot et al. [11] who investigated a similar local, cross-flow displacement ventilation for conveyor
belt air conditioning, calculated total energy savings of up to 32 %. In the case of application, the
local air conditioning of sandwiches requiring a supply temperature of 5-10 °C was investigated. The
supply air speeds were varied between 0.17 and 0.51 m/s in this study. Due to the local conveyor belt
climate control, the hall temperature was raised to 16 °C and a relative humidity of 55 % was set in

T
the calculations by Burfoot et al [11]. The comparatively lower energy savings can therefore be

IP
explained by the lower supply air temperatures for conveyor belt air conditioning and hall air
conditioning.

CR
6. Conclusion and outlook

The laboratory experiments as well as the simulation calculations prove, that the displacement

for preventing the infiltration of water droplets. US

ventilation is very well suited for keeping temperature limits and limits of relative humidity, but not

Under the favored parameter combination (supply air temperature: 16 °C, outlet velocity: 0.1 m/s,
AN
outlet ratio 1:8), high energy savings can be realized for the application. Likewise, a sufficiently
stable temperature profile and a sufficiently high product safety are ensured. In addition, a free access
to the product is possible.
However, there is an increased effort to clean the outlets and the conveyor belt and there are also
M

increased installation costs.

It should be noted, that the local air conditioning via a cross-flowing displacement ventilation as a
ED

further development of a full encapsulation of the product-carrying conveyor belt offers significant
advantages. An optimal combination of parameters could be determined for the application, a
practical design proves the functionality.
As described, the implementation of a local product air conditioning system enables the expansion of
PT

the temperature and humidity limits in the hall and thus increases the flexibility of the HVAC mode of
operation. To be able to use this flexibility also for load management or to enable operation via
flexible electricity prices, a model predictive control of the HVAC is in progress. The advantages of
CE

using a model predictive control for HVAC systems has already been described in many ways [23-
25]. In this model predictive control weather forecasts, production plan dependent, internal loads of
the hall and electricity price forecasts will be implemented. Using state space models that map the
AC

future behavior of the system and optimization calculations, the forecast data is processed into optimal
control variables for the HVAC, optimizing flexible operation [6].

Acknowledgements

The contents of this paper have been acquired within the cooperation project „Smart Consumer –
Energy efficiency through systemic coupling of energy flows by means of intelligent measurement
and control technology”. The Project is funded by the German Federal Ministry for Economic Affairs
and Energy (FKZ: 03ET1180).
 ACCEPTED MANUSCRIPT

Nomenclature

∇ Nabla operator [-]

ρ Density [kg/m³]
Cp Isobaric specific heat capacity [J/K]
ε Thermodynamic internal energy [J]
KH Heat conduction coefficient [W/m K]
P Pressure [N/m²]
R² Coefficient of determination [-]
T Temperature [K]
t Time [s]
u Velocity [m/s]

T
References

IP
[1] G. Guyot, M.H. Sherman, I.S. Walker, Smart ventilation energy and indoor air quality performance in

CR
residential buildings: A review, Energy and Buildings 0000 (2018) 1-15,
doi:10.1016/j.enbuild.2017.12.051 .
[2] R. Hitchin, C. Pout, D. Butler, Realisable 10-year reductions in European energy consumption for air

US
conditioning. Energy and Buildings 86 (2015) 478-491, doi:10.1016/j.enbuild.2014.10.047 .
[3] K. Zhang, Z. Xiaosong, L. Shuhong, J. Xing, Review of underfloor air distribution technology, Energy
and Buildings 85 (2014) 180-186, doi: 10.1016/j.enbuild.2014.09.011 .
AN
[4] J. Wagner, 2016, Local air conditioning of temperature-sensitive products - Experimental
investigations and energetic potential analysis, Dissertation, Kassel University Press.
[5] M. Schäfer, J. Wagner, J. Hesselbach, C.X. Lin, Comparison of mixed and stratified ventilation
systems in production areas – experimental and simulation studies, Proceedings of the ASME 2013
M

Summer Heat Transfer Conference, July 14-19, 2013. Minnesota, USA, doi:10.1115/HT2013-17646 .
[6] T. Heidrich, H. Dunkelberg, T. Weiß, J. Hesselbach, 2017, Flexibilization of Energy Supply Using the
Example of Industrial Hall Climatization and Cold Production, Simulation in Production and Logistic,
ED

pp. 39-48, doi: 10.19211/KUP9783737601931.

[7] T. Heidrich, Energieeffizienzsteigerung in der Süßwarenindustrie durch lokale Klimatisierung und
systemische Kopplung von Energieströmen. Talk at Energy Decentral – DLG (German Agricultural
PT

Society) session - innovative energy technologies in the food industry, November 17, 2016,
Hannover, Germany, doi: 10.13140/RG.2.2.14958.92486 .
[8] J. Wagner, M. Schäfer, A. Schlüter, L. Harsch, J. Hesselbach, M. Rosano, L. Cheng-Xian, Reducing
CE

energy demand in production environment requiring refrigeration—A localized climatization

approach. HVAC&R Research Vol. 20, Iss. 6 (2014) 628-642, doi: 10.1080/10789669.2014.929451 .
[9] L. Olander, L. Conroy, I. Kulmala, R.P. Garrison, M. Ellenbecker, B. Biegert, B. Flecther, H.D.
AC

Goodfellow, G. Rosen, B. Ljungqvist, B. Reinmüller, A. Dumas, M. Robinson, 2001, Industrial

Ventilation Design Guidebook - Chapter 10 – Local Ventilation. Academic Press, San Diego, USA,
doi:10.1016/B978-012289676-7/50013-8 .
[10] D. Burfoot, K. Brown, Y. Xu, S.V. Reavell, K. Hall, Localised air delivery systems in the food
industry Volume 11 (2000) 410-418, doi.org/10.1016/S0924-2244(01)00021-8 .
[11] D. Burfoot, S. Reavella, D. Wilkinson, N. Duke, Localised air delivery to reduce energy use in the
food industry, Journal of Food Engineering Volume 64 (2004) 23-28, doi:10.1016/S0260-
8774(03)00166-3.
[12] M. Murakami, S. Aizawa (1999), Belt Conveyor Device for Transporting Cold Foods, PCT Patent
Application WO 99/50605.
 ACCEPTED MANUSCRIPT

[13] A. Q. Ahmed, S. Gao, A. K. Kareem, Energy saving and indoor thermal comfort evaluation using a
novel local exhaust ventilation system for office rooms, Applied Thermal Engineering 110 (2017)
821–834, Elsevier, doi:10.1016/j.applthermaleng.2016.08.217 .
[14] D. Al Assaad, K. Ghalia, N. Ghaddara, C. Habchiba, Mixing ventilation coupled with personalized
sinusoidal ventilation: Optimal frequency and flow rate for acceptable air quality Energy and
Buildings 154 (2017) 569–580, doi:10.1016/j.enbuild.2017.08.090 .
[15] B. Xia, D.-W. Sun, Applications of computational fluid dynamics (CFD) in the food industry: a
review, Computers and Electronics in Agriculture 34 (2002), pp. 5–24, Elsevier, doi:10.1016/S0168-
1699(01)00177-6 .
[16] H. Montazeri, F. Montazeri: CFD simulation of cross-ventilation in buildings using rooftop
windcatchers: Impact of outlet openings, Renewable Energy 118 (2018) 502-520,

T
doi:10.1016/j.renene.2017.11.032 .

IP
[17] D. Parpas, C. Amaris, S.A. Tassou, Investigation into air distribution systems and thermal
environment control in chilled food processing facilities, international journal of refrigeration (2017),

CR
doi: 10.1016/j.ijrefrig.2017.10.019
[18] Y.A. Tabunshchikov, V.E. Shkarpet, D.V. Kapko, M.M. Brodach, K.V. Kochariantc, Installing
Local Recirculation Air Diffusers during Building Deep Renovation Reduces Energy Consumption of
Ventilation Systems, SBE 16 Tallinn and Helsinki Conference; Build Green and Renovate Deep, 5-7

doi:10.1016/j.egypro.2016.09.149 . US
October 2016, Tallinn and Helsinki, Energy Procedia 96 (2016) 266 – 276, Elsevier,

[19] Q. Chen, Ventilation performance prediction for buildings: A method overview and recent
AN
applications, Building and Environment 44 (2009) 848–858, Elsevier,
doi:10.1016/j.buildenv.2008.05.025 .
[20] VDI Guideline 2078, Cooling Load Calculation of Air-conditioned Rooms, Association of German
Engineers. 2012.
M

[21] M. Schäfer, J. Wagner, A. Schlüter, J. Hesselbach, Energy Flows in Industrial Buildings and
Machines using the Example of a Node Model, Simulation Notes Europe 22 (2012) 1-6, doi:
ED

10.11128/sne.22.tn.10103 .
[22] S. Schirmer, T. Heidrich, J. Wagner, J. Hesselbach, Steigerung der Energieeffizienz einer
klimatisierten Produktionshalle, HLH - edition July/August (2016) 38-40, Publisher VDI-Springer.
]23] A. Afram, F. Janabi-Sharifi, Theory and applications of HVAC control systems – A review of model
PT

predictive control (MPC). Building and Environment 72 (2014) 343-355, doi:

10.1016/j.buildenv.2013.11.016 .
[24] J. Alvarez, J.L. Redondo, E. Camponogara, J. Normey-Rico, M. Berenguel, P.M. Ortigosa,
CE

Optimizing building comfort temperature regulation via model predictive control, Energy and
Buildings 57 (2013) 361-372, doi:10.1016/j.enbuild.2012.10.044 .
[25] F. Oldewurtel, A. Parisio, C.N. Jones, D. Gyalistras, M. Gwerder, V. Stauch, B. Lehmann, M.
AC

Morari, Use of model predictive control and weather forecasts for energy efficient building climate
control. Energy and Buildings 45 (2012) 15-27, doi:10.1016/j.enbuild.2011.09.022 .