Wyniki 1-8 spośród 8 dla zapytania: authorDesc:"Paolo DI BARBA"

A multiobjective approach to the shape design of HTSC magnetic bearings

Czytaj za darmo! »

The shape design of the excitation system of a high-temperature-superconductor (HTSC) magnetic bearing is revisited in terms of multiobjective design. It is proven how the concept of non-dominated solution, helps the designer to represent the feasible region in the objective space and identify its boundary. At the end of the proposed design procedure, a set of optimal geometries of the device - [...]

Zastosowanie programu Remcom XFdtd i środowiska Octave do optymalizacji anten DOI:10.15199/59.2015.4.7


  Projektowanie takich anten, dla których nie istnieją proste wzory obliczeniowe, jest przeważnie złożonym zadaniem optymalizacyjnym. Obejmuje ono zarówno wybór kształtu (typu) anteny, jak i dobór wartości parametrów geometrycznych oraz materiałowych, tak aby zostały spełnione wymagania specyfikacji, np. odpowiednio ukształtowana charakterystyka promieniowania, dopasowanie impedancyjne. W procesie optymalizacji wykorzystuje się programy komputerowe z zakresu elektrodynamiki obliczeniowej, służące do wyznaczenia funkcji celu, która jest pochodną parametrów anteny. W przypadku anten, których projektowanie wymaga wyznaczenia wartości wielu parametrów geometrycznych czy materiałowych, jednym z teoretycznie możliwych podejść jest przeprowadzenie optymalizacji metodą wyczerpującego przeszukiwania. Polega ona na symulacji właściwości anteny dla wszystkich kombinacji wartości parametrów w ich zdyskretyzowanej przestrzeni i wyborze wariantu optymalizującego wartość funkcji celu. Już przy stosunkowo niewielkiej liczbie parametrów jest to zwykle zadanie bardzo (lub wręcz zaporowo) czasochłonne, co wynika z dużego nakładu obliczeniowego potrzebnego do wielokrotnego wykonania symulacji elektrodynamicznych. Proces projektowania może zostać znacznie skrócony, dzięki rezygnacji z tego algorytmu na rzecz heurystycznych metod optymalizacji. Nie dają one gwarancji znalezienia rozwiązania optymalnego, ale umożliwiają znacznie szybsze przeszukiwanie przestrzeni parametrów. Zapewniają one duże prawdopodobieństwo znalezienia w rozsądnym czasie rozwiązania, które jest akceptowalne z punktu widzenia przyjętych założeń projektowych. Dotychczas opracowano i stosowano wiele algorytmów optymalizacji anten, co świadczy ogólnie o skuteczności takiego podejścia do projektowania [2 - 4]. Dodatkowo w ostatnich latach, dzięki rozwojowi technologii komputerowej skutkującej znacznym przyspieszeniem obliczeń, możliwe było wykorzystanie w praktyce projektowej meto[...]

AUTOMATYCZNA IDENTYFIKACJA PARAMETRÓW UPROSZCZONEGO MODELU CIAŁA CZŁOWIEKA ZA POMOCĄ ALGORYTMU EWOLUCYJNEGO DOI:10.15199/59.2016.6.94


  AUTOMATED IDENTIFICATION OF HUMAN-BODY MODEL PARAMETERS WITH EVOLUTIONARY ALGORYTHM Streszczenie: W artykule przedstawiono wykorzystanie algorytmu ewolucyjnego (Estra) do identyfikacji parametrów uproszczonego modelu ludzkiego ciała (fantomu). Uproszczony model ciała może być wykorzystywany do symulacji odstrojenia impedancyjnego anteny znajdującej się w pobliżu ciała. W artykule przedstawiono sposób określenia parametrów uproszczonego modelu za pomocą automatycznej procedury opartej na algorytmie ewolucyjnym i metodzie różnic skończonych w dziedzinie czasu (FDTD). Po określeniu wartości parametrów, uproszczony model został porównany do heterogenicznego modelu ludzkiego ciała. Modele porównano w oparciu o analizę dopasowania impedancyjnego anteny dipolowej znajdującej się na obu modelach. Abstract: The paper presents the exploitation of a lowestorder algorithm of evolutionary computing (EStra) for identifying the parameters of a simplified human body model (phantom). A simplified model is well suited in view of the computationally-expensive field simulation of wearable antennas located in a close proximity to the human body. In the paper, an automated procedure based on evolutionary computing and Finite Difference Time Domain (FDTD) computational electrodynamics method is proposed to identify the parameters of the simplified model. Subsequently, after identifying the parameter values, the simplified model is compared to a heterogeneous anthropomorphic human-body model. The comparison is based on the analysis of impedance matching of the same dipole antenna located on both the anthropomorphic and simplified phantoms . Słowa kluczowe: identyfikacja parametryczna, algorytmy ewolucyjne, FDTD, modele ciała człowieka, anteny nasobne. Keywords: Parameter identification, evolutionary computing, finite-difference time domain, human body model, wearable antennas. 1. UPROSZCZONY MODEL CIAŁA CZŁOWIEKA Projektowanie i optymalizacj[...]

DWUKRYTERIALNE EWOLUCYJNE PODEJŚCIE DO OPTYMALIZACJI ROZMIESZCZENIA WĘZŁÓW DOSTĘPOWYCH DLA POPRAWY ZASIĘGU I DOKŁADNOŚCI LOKALIZACJI TERMINALI DOI:10.15199/59.2019.6.81


  1. INTRODUCTION Nowadays, radio networks are widely used not only for the provision of fundamental communication services, such as voice and data exchange but also for the provision of multitude of value added applications. With the rapid growth of the Internet of Things (IoT) market, densification of radio access networks increases the significance of network planning and optimization. The radio access network has to satisfy multiple criteria to provide reliable and energy efficient services in a wide range of propagation environments for a variety of terminals, keeping intra-network interference as low as possible. The key services of interest are terminal positioning and user mobility patterns. Among most frequently used positioning methods are those using received signal strength (RSS) for the estimation of terminal location [1][2][3][4]. The accuracy of signal strength based positioning is strongly affected by the complexity of radio wave propagation in indoor multipath environments. Thus, it is vital to design the network in a way, that allows to minimize the errors resulting from relative positions (geometry) of terminals and reference transmitters (often referred to as radio beacons) or access points (AP). It is expected that small errors in the measured signal levels should result in small positioning errors [5]. In literature, there are many examples of the application of evolutionary optimization algorithms to simultaneous improvement of coverage, reliability, energy-limited lifetime, deployment cost, and other parameters of radio access networks [6][7][8]. In this paper we focus on the enhancement of an evolution-inspired network optimization method which simultaneously takes into account both the coverage and positioning accuracy for 2-D deployment of indoor access points, previously described in [9]. We chose evolutionary approach to our optimization problem which is explained in section 2. The remainder of[...]

EWOLUCYJNE PODEJŚCIE DO AUTOMATYCZNEGO ROZMIESZCZANIA PUNKTÓW DOSTĘPOWYCH WLAN DO JEDNOCZESNEJ POPRAWY ZASIĘGU I DOKŁADNOŚCI LOKALIZACJI DOI:10.15199/59.2017.6.33


  Nowadays, Wireless Local Area Networks (WLANs) are pervasive and are more and more frequently used not only for internet access but also for indoor location services. A number of indoor positioning approaches and systems have been developed, which mostly use received signal strength indicator (RSSI) or time difference of arrival (TDoA) methods. Although many problems related to the development of wireless indoor positioning systems have been addressed and already reported in the literature [2, 3, 4, 8, 9, 11], their accuracy is restricted not only by the complexity of radio wave propagation in multipath environments but also by the fundamental limitations of multilateration techniques which follow from the relative position (geometry) of terminals and beacons (in the considered case WLAN access points). Preferably, (small) errors in the measured RSSI/TDoA data should not result in large positioning errors. This problem has been extensively addressed e.g. in [5, 6] and is illustrated in Fig. 1. AP1 AP2 AP1 AP2 Fig. 1. Simple illustration of the dependence of the positioning error on transmitter-receiver geometry: good geometry (left), poor geometry (right) In this paper we focus on the development of a novel evolution-inspired improvement method which simultaneously takes into account both power coverage and positioning accuracy for the deployment of WLAN access points in indoor environment. The remainder of the paper is organized as follows: Section 2 describes the deployment method, Section 3 contains outcomes of the proposed method and Section 4 summarizes the results. 2. METHOD In this research we aim at simultaneous improvement of coverage and indoor positioning accuracy with the number of nodes (access points/ beacons) as small as possible. In the following subsections we will define an objective function and the improvement (optimization) method involving evolutionary computing. 2.1. Criteria for the Placement[...]

Automated optimal design of wells for electromagnetic cell stimulation DOI:10.15199/48.2019.05.01

Czytaj za darmo! »

In the last decades, the electromagnetic stimulation in vitro and in vivo has become a promising research field because it allows to modulate the behaviour of cells and tissues. In particular, when the cells are exposed to a timevarying magnetic field, an electric field is induced and thus a current density arises, because the cell culture medium is conductive. The interaction between the induced current density and the time-varying magnetic field gives rise to mechanical stress acting on the cells [1]. In this paper, new kind of wells for obtaining a homogeneous stress and stimulation of a considerable large amount of cells are designed [2]. This design problem is formulated as a multiobjective one and its solution is found by means of the Biogeography-Inspired Multi-objective Optimization algorithms, BiMO [3,4] and the μ-BiMO algorithms [5]. These methods have shown to be successful for various applications [6-10]. In particular, when the forward problem requires a high computational time e.g. when Finite Element FE simulations are used, the μ-BiMO algorithm gave good results. In general, the aim of this study has been to design different optimally-shaped wells for electromagnetic stimulation of cells [11-15]. The forward problem The electromagnetic stimulation of cells is done by means of the so-called “electromagnetic bioreactor" (Fig. 1), which is a device based on two solenoids connected in series and powered by a pulse generator (Igea, Carpi, Italy) at 75 Hz [11]. In order to simulate the electric E and magnetic B fields in the bioreactor, a 3D time-dependent finite-element model was implemented in MagNet, a commercial code by Mentor-Infolytica. Fig.1. Electromagnetic bioreactor Fig.2. Magnetic induction field [T] distribution in the middle of the bioreactor In the conductive regions, the electromagnetic problem is solved in terms of the phasors of the electric vector potential T and the scalar ma[...]

Field models of induction heating for industrial applications DOI:10.15199/48.2018.03.01

Czytaj za darmo! »

In the community of computational electromagnetics, the set of benchmark problems proposed by the TEAM (Testing Electromagnetic Analysis Methods) series of workshop is a reference for testing numerical methods in a comparative way [1]-[2]. Nevertheless, there is a lack of problems specifically focused on induction heating devices, as far as numerical modelling is concerned. More generally, in the past some benchmarks of induction heating was proposed, but the attention was focused rather on the inverse problem [3]-[7] related to the optimal design of the power inductor than on the direct problem of field analysis [8], [9]. In fact, in computational induction heating, analysis problems are challenging because they involve different physical domains; therefore, the development of non-linear coupledfield models and the consequent choice of suitable solvers is mandatory [8]-[10]. Too often numerical solvers, like e.g. finite-element solvers which are commercially available, are used by designers as general-purpose black boxes. Moving from this background, it was proposed to define a benchmark of coupled-field analysis [5]; the problem is taken from industrial applications of induction heating: it deals with the transient thermal analysis of a steel-made cylindrical billet, subject to the changing magnetic field of a multi-turn winding. It is a clear example showing that stiff analysis problem can originate even in the case of very simple geometries. Benchmark description: the device The device under study is composed of an inductor winding and a cylindrical billet; winding and billet are coaxially located. A. Geometry The billet has a radius r and height h. The inductor is made of 20 hollow circular turns, connected in series; each of them has height hc and width wc, while their radial distance from Y axis is rc. The thickness of the copper of each hollow turn is tc. Numerical data about the geometry are summarized in Table [...]

ViMeLa Project: An innovative concept for teaching mechatronics using virtual reality DOI:10.15199/48.2019.05.05

Czytaj za darmo! »

Nowadays, traditional education and teaching methods, although with significantly improved teaching techniques, can not keep enough interest of the students that grew up with Internet, mobiles and tablets. Especially sensitive to these issues are students in engineering, in particular, in mechatronics. Modern information technology is rapidly being adopted in Mechatronics Engineering education as a tool for enriching the practical experience of the students. The practical training is a vital part of Mechatronics Engineering education [1]. However, the high cost needed to implement laboratory experiments (for educational purposes) led to development of virtual facilities where physical systems can be virtually controlled via the Virtual Reality (VR) simulations. Multimedia and VR technologies offer great potential for presenting theory and laboratory experiments in an enhancing and interesting, but in an economical, way. Teaching and learning Mechatronics Mechatronics is synergy and interaction of mechanical, electrical and computer systems as seen in Fig. 1. Hence, it is an interactive combination of mechanical engineering, electronic control and computer technology, with the aim of achieving an ideal balance between mechanical structure and its overall control and performance. Fig.1. Structure and key elements of mechatronics Currently, mechatronics classes are divided into two parts: the theoretical lectures and laboratory courses with experiments following the "learning by doing" model. Expensive equipment and limited time for training do not provide sufficient educational platforms [2,3]. In some cases the students conduct based simulations and learn how mechatronic systems and devices operate in reality, despite it may seem abstract and unclear for students, and does not fully reflect the physical phenomena of particular processes. The described drawbacks of mechatronics study are greatly improved when classroom teac[...]

 Strona 1