15:40
Session 4: Engineering Applications
Chair: Henk Polinder
15:40
20 mins
|
A MODIFIED MIGRATION MODEL BIOGEOGRAPHY EVOLUTIONARY APPROACH FOR ELECTROMAGNETIC DEVICE MULTIOBJECTIVE OPTIMIZATION
Mahmoud Sayed, Ahmed Abdallh, Luc Dupré
Abstract: In this paper, we present an efficient and robust algorithm for multiobjective optimization of electromagnetic devices. The recently developed biogeography-based optimization (BBO) is modified by
adapting its migration model function so as to improve its convergence. The proposed Modified Migration Model biogeography-based optimization (MMMBBO) algorithm is applied into the optimal geometrical design of an electromagnetic actuator. This multiobjective optimization problem is solved by maximizing the output force as well as minimizing the total weight of the actuator. The comparison between the optimization results using BBO and MMMBBO shows the superiority of the proposed
approach.
|
16:00
20 mins
|
MAXIMIZATION OF TEMPERATURE BUILD-UP IN BIPOLAR RFA USING PULSED VOLTAGE PROFILES
Frederik Soetaert, Guillaume Crevecoeur, Luc Dupré
Abstract: Radiofrequency ablation (RFA) is an auspicious alternative cancer treatment that creates heat by means of electrodes that are subject to radiofrequency voltage differences. We propose the use of a bipolar RFA device, with two electrodes, that applies a pulsed voltage profile. To understand the physical phenomena to a better extent, we have implemented a 3D finite element model of pulsed bipolar RFA. We furthermore propose a numerical optimization of the pulsed voltage profile in order to obtain a temperature build-up in tumor regions that normally survive RFA. Numerical results confirm that this optimization results in a larger temperature build-up and thus increases the efficacy of the treatment.
|
16:20
20 mins
|
OPTIMAL PROBE DESIGN FOR LORENTZ FORCE EDDY CURRENT TESTING
Konstantin Porzig, Reinhard Schmidt, Marek Ziolkowski, Matthias Carlstedt, Hartmut Brauer, Hannes Toepfer
Abstract: Lorentz force eddy current testing is a contactless method to investigate conductive materials regarding the presence of inclusions and defects. The performance of such a system strongly depends on the applied probe which acts as an electromagnetic field source and sensor at the same time. We propose an optimization strategy to maximize the absolute defect response signal while ensuring technical feasibility. The present study shows that cylindrical magnets should be preferably used for deep lying subsurface defects while Halbach cylinders outperform the latter in case of surface flaws due to a more focal magnetic field.
|
16:40
20 mins
|
OPTIMIZATION OF ENERGY STORAGE SCHEDULING IN ENERGY HUBS
Arianna Ghezzi, Giambattista Gruosso, Maurizio Repetto
Abstract: A energy hub is a local energy system where generation, loads and storage are interacting among them and with external infrastructure. The power production scheduling of the various modules can be handled by an optimization procedure that tries to minimize the overall energy cost on a given time period, for instance one day. In this paper a new kind of shape functions is used to model the power going in and out of the storage system. These functions satisfy some basic properties of the storage and reduce the number of degrees of freedom of the optimization. The new shape functions are described and some examples, showing their advantages, are presented.
|
17:00
20 mins
|
A GENERIC INVERSE PROBLEM ALGORITHM FOR THE IDENTIFICATION OF THE MAGNETIC MATERIAL PROPERTIES OF ELECTROMAGNETIC DEVICES COUPLED WITH STOCHASTIC UNCERTAINTY ANALYSIS
Ahmed Abdallh, Luc Dupré
Abstract: Magnetic characteristics of the core materials of electromagnetic devices are retrieved by solving a combined experimental-numerical electromagnetic inverse problem. However, both measurement noise and uncertainties of the forward model may deteriorate the accuracy of the recovered solution. In this perspective, we review the use of the electromagnetic inverse problem for the identification of the magnetic material properties. The inverse algorithm is combined with a stochastic uncertainty analysis for a priori qualitative error estimation and a quantitative error reduction. The complete inverse methodology is applied onto the identification of several macroscopic properties of the magnetic material of a wide range of rotating electrical machines. Numerical and experimental validations of the proposed approach are presented. This complete inverse algorithm can be applied into a wide range of applications in electromagnetic community.
|
|
