ABSTRACTS
Computer
Simulation of Ice Cream Solidification
Authors J.
Aldazábal, A. Martín-Meizoso, J. M. Martínez-Esnaola,
R. Farr
Speaker Dr.
Javier Aldazábal
Centro de Estudios
e Investigaciones Técnicas de Gipuzkoa (CEIT), Spain
Abstract This
work proposes three different models to simulate the evolution of ice cream
microstructure during the manufacturing process. The simulated process
corresponds to the solidification in the blast freezer of ice seeds that
have been generated in the barrel freezer. All these models are implemented
in several computer codes to study their behaviours and predictions. The
first model developed here works with an analytical description of the
microstructure. It considers the ice particles as spheres. Their evolution
is simulated by increasing the size of particles and controlling the dihedral
angles that appear among them. The second model proposed is based on the
Monte Carlo technique. This model simulates the evolution of a discretized
version of the microstructure obtained from the barrel freezer. It includes
the sugar concentration in the matrix and its diffusion. The last model
is deterministic. It works with a voxelised version of the initial microstructure.
The evolution of the microstructure is simulated by solidifying isotropic
ice layers surrounding initial solid particles. These layers are grown
considering the sugar concentration in the liquid-matrix. This method is
able to quantitatively relate the simulated and real time. |
Modeling
and Control of the olive oil extraction process
Author C.
Bordons
Speaker Carlos
Bordons Universidad de Sevilla, Spain
Abstract The
automatic control of the extraction of oil out of olives is still an open
field, since many installations are usually operated in manual mode. As
olive oil mills are becoming bigger the chances for automation are increasing,
therefore it is important to acquire the necessary knowledge of the process
behaviour in order to design the appropriate control strategies. This talk
deals about some issues related to the modelling and automatic control
of olive oil mills. After a brief review of the state of the art in automatic
control of oil mills, two main topics are addressed: measurements of crucial
variables and automatic control of olive oil yield. Related to the first
topic, the concept of soft sensor is introduced and it is shown how it
can be used in the olive oil elaboration process. In an olive oil
mill, soft sensors measure variables that in other case should come either
from laboratory analysis with large processing delays or from very expensive
and difficult to use on-line analyzers. The introduction of these sensors
in the olive oil industry can help to increase the degree of automation
in the elaboration process. The sensors have been devised based upon artificial
Neural Networks (NN) and have been successfully tested on a real mill.
A solution to the integrated control of the mill is also presented, where
a predictive strategy is used to optimize oil yield while keeping quality
standards. The talk also shows the application problems that arise when
implementing advanced controllers in an industrial control system with
low computational capabilities. The talk will also deal with the main issues
related to integration and will show some results of the application of
advanced control techniques to an actual olive oil mill. The application
of the proposed strategies has shown that great benefits can be obtained
both in oil yield and extraction performance. |
Travelling
wave solutions for a model of a fungal disease of the vine
Authors J.
B. Burie, A. Calonnec, A. Ducrot
Speaker Dr.
Jean-Baptiste Burie
Université de Bordeaux,
France
Abstract Powdery
mildew, caused by the fungus Uncinula necator, is the most economically
important and widespread disease of grapevines. A better knowledge of its
propagation mechanisms could help to improve the treatment strategies.
In this talk, we will consider two mathematical models describing the epidemic
at the vineyard scale. They both include reaction-diffusion equations that
modelize the dispersal in the vineyard of the spores produced by the fungus.
The evolution of the fungus is either modelized
by a delayed differential
equation (first model, Zawolek & Zadoks), or by a set of ODEs of SEIR-type
(second model). Both models include the main biological characteristics
of the disease: multiple ranges for spores dispersal, a latent and an infective
stage for the fungus. We will prove existence and unicity (up to translations)
of 1D travelling waves for both models if the speed of the travelling wave
is larger than some minimal speed c*. A formula for the computation of
the speed in terms of the various parameters is derived. Numerical simulations
confirm the existence of a travelling wave with a speed equal to the minimum
speed c*. Replacing the various biological parameters of the model with
realistic estimations, we find that there is a good match between c*and
the velocity of the epidemic observed by the biologists on the field. We
will also study the qualitative dependence on the biological parameters
of the velocity c*. We will finally give some computation results in 2D
that show the role of the row structure of the vineyards on the spread
of the epidemics. |
Inverse
Problems Arising in the Modelling of Food Processes
Authors A.
Fraguela, J. A. Infante, Á. M. Ramos, J. M. Rey
Speaker Dr.
Andrés Fraguela
Benemérita Universidad
Autónoma de Puebla, México
Abstract In
our talk we propose a stable method for the identification of the diffusion
coefficient in a mathematical model that describe the phenomena of heat
transfer taking place during the high-pressure treatment of foods. In a
simple version of the model we solve an inverse problem for the identification
of a pressure depending diffusion coefficient in a boundary value problem
of parabolic type starting from measurements of the temperature at a given
location of the considered spatial domain during certain interval of time.The
pressure dependence of the diffusion coefficient is the main complication
in the solution of this inverse problem. |
Influence
of High Hydrostatic Pressure on Thermofuiddynamics of Pressurized Food
Authors W.
Kowalczyk, A. Delgado
Speaker Dr.
Wojciech KowalczykUniversity
Erlangen-Nürnberg, Germany
Abstract The
treatment of food with high hydrostatic pressure (HHP) up to 10.000 bars
allows an increase of food quality and a diversification of available products.
Both experimental and numerical results show that HHP influences significantly
fluid flow phenomena occurring in autoclaves. In the current contribution,
heat and momentum transfer during high pressure treatment of liquid substances
is discussed. In order to characterise the thermofluiddynamics in high
pressure chambers, the dimensional analysis of the governing equations
is carried out. Both cases with free and with forced convection as well
as examples accompanied by phase transitions are considered. The dependency
of convective and diffusive heat transfer on temperature and pressure is
presented. Additionally, the heat and momentum transfer in high viscous
media and chambers with different characteristic length is analysed. |
Modeling
the risk of entrance of Foot-and-mouth disease (FMD) in Spain
Authors J.
M. Sánchez-Vizcaíno, B. Martínez, M. Martínez
Speaker Dr.
J. M. Sánchez-Vizcaíno
Universidad
Complutense de Madrid, Spain
Abstract Foot-and-mouth
disease (FMD) is a highly infectious animal disease which affects many
species of cloven-hoofed animals and causes important economic loses. Although
FMD was successfully eradicated in the European Union in 1992, occasional
incursions have been occurred being the most important in 2001, where United
Kingdom (UK), Ireland, France and Netherlands were affected. In order to
assess the risk of FMD introduction in Spain we have developed a stochastic
model which includes all the potential routes of the disease entrance.
Our objective was to determine the overall risk of FMD introduction in
Spain, the provinces with more risk and the most probable routes of entrance.
In this lecture we will describe the considered scenarios in order to understand
the sequence of events in time and the interactions among them. The model
includes parameters related with the virus, susceptible animals, high risk
materials and the probability of exposure and contact. Probability distributions
were used for uncertain variables and stochastic effects were obtained
using the Monte Carlo simulation (@Risk). The model not only provides the
high risk Provinces and the most probable routes of entrance, but also
a way to quantify the effectiveness of a control measures implementation.
This information can be used to increase the surveillance and control measures
in the critical points to reduce as much as possible the probability of
introduction
of the FMD into the country. |
Heat
and Mass Transfer modelling in High-Pressure Food Processing
Authors P.
D. Sanz, L. Otero, Á. M. Ramos, C. de Elvira, J. S. Torrecilla
Speaker Dr.
Pedro D. Sanz
CSIC, Spain
Abstract High-pressure
processing is an expanding technology that is being successfully introduced
in the food industry mainly as a preservation method to control the microbiologic
and enzymatic activities in food (Torres and Velazquez, 2005). Another
promising application of high-pressure technology, nowadays unexploited
but under active research, includes processes involving phase transitions:
high-pressure freezing and thawing. Pressure allows the displacement along
the phase diagram of water and different kinds of high-pressure freezing
and thawing processes can be performed (Urrutia Benet et al., 2004).
Among them, high-pressure assisted freezing has a special interest in that
it allows obtaining different ice polymorphs depending on the pressure
level at which the freezing process has been developed. All ice polymorphs
different than ice I are denser than liquid water; therefore, a decrease
in volume occurs during the phase transition and low damage is caused to
the sample structure (Molina-García et al., 2004; Luscher
et
al., 2005). In this study an overview of the fundamental, current and
future applications of the high pressure is presented. Also a mathematical
model of the behaviour of thermophysical properties is presented as well
as other mathematical model is described to account for the phenomena of
heat and mass transfer taking place during pressurization of a food model.
High-pressure experiments were carried out in agar gel samples. During
all the experiments, the temperature evolution in the centre and in the
surface of the samples were recorded. The mathematical model took
into account the free convection that arises in the pressure medium after
compression because it has been shown that it affects the thermal evolution
of the processed food in a significant way (Otero et al., 2006).
Simulations made with the finite element method showed good
agreement between experimental and predicted temperature profiles for all
the processes studied. Modelling of high-pressure processing is important
for the optimization of these new processes and their general implantation
in the food industry. Also it is presented a numerical model to simulate
Pressure-Shift Freezing processes in a tubular reactor designed to work
in a semi-continuous mode on an industrial scale. The model will make possible
to optimize different parameters such as the length of the pressure vessel
and the displacement rate of the sample. |
Multiscale
modelling of transport phenomena in fruits
Authors Pieter
Verboven, Ho Quang Tri, Hibru Mebatsion, Trung Anh Nguyen, Fernando Mendoza,
Bart Nicolaï
Speaker Dr.
Pieter Verboven
Katholieke Universiteit Leuven,
Belgium
Abstract Transport
of gasses and moisture are crucial for maintaining fruit quality during
and after storage. These phenomena have been studied and modelled by assuming
the product to be composed of different continuous materials (cuticle,
outer cortex, inner cortex, vascular tissue). The model parameters, such
as the mass diffusion coefficients of the tissue, should however be considered
as apparent material parameters which incorporate both actual physical
constants such as the properties of water and air but also the microscale
geometry of the tissue and the intracellular space. The relationship between
the macroscopic apparent properties and the microscopic features is not
understood well to date. As a consequence, the available continuum models
have a limited range of validity. A multiscale modelling approach is more
appropriate than a macroscopic continuum modelling approach to understand
the relative importance of microscopic features on the overall behaviour
of the biological product. In the other direction, multiscale models will
assist the determination of microscopic triggers for internal damage as
a consequence of macroscopic environmental changes. Multiscale models are
basically a hierarchy of submodels which describe the behaviour at different
spatial scales in such a way that the submodels are interconnected. This
paper will reveiw the recent advances made in this field. |
