National Technical University of Athens (NTUA)
Ανάπτυξη μεθόδων μέτρησης και ανάπτυξη αλγορίθμων μοντελοποίησης και ανάλυσης μαγνητικής συμπεριφοράς συσκευών για επίτευξη μαγνητικής καθαρότητας σε διαστημικές αποστολές
Abstract
dc:descriptionThe present PhD thesis aims at the development of stochastic algorithms for modeling and analyzing the magnetic behavior of equipment for magnetic cleanliness study in space missions. Additionally, several parameters of the magnetic field’s measuring procedure of Equipment under test (EUT) is studied, since the modeling of an EUT is based on the measured magnetic field it produces.At first, in chapter 1, the importance of the magnetic cleanliness in space missions is explained with reference to older space missions where it was attempted. Moreover, the procedure and the equipment that is been used for magnetic field measurements in certified facilities are presented along with the approach used for extracting models with equivalent magnetic behavior to the EUTs in a spacecraft.In chapter 2 the uncertainty in the measuring process is presented, analyzing its 2 types and the way to approach and calculate them with an indicative application of identifying uncertainty factors in the measuring process of a magnetic field in a certified facility.In chapter 3, the two stochastic optimization methods, PSO and GA, which are used in the present thesis for the development of the algorithms that will produce the equivalent models of EUTs by using the magnetic field measurements it produces, are presented.Chapter 4 presents the first stochastic algorithm, based on PSO, for modeling EUT. In this chapter the problem and the approach this thesis follows are described and the results that are produced for simulated EUTs are presented, confirming the algorithms performance.Afterwards, in chapters 5 and 6, two algorithms are developed, based on the approach in chapter 4, that are focused on a calibration method for magnetic field’s measuring facilities and on the modeling of EUT using simultaneously a dipole and a quadrupole to comprise the model respectively. Their performance is confirmed from the results that are produced and also presented in the chapter.Chapter 7 includes the second stochastic algorithm for modeling magnetic fields, in which the number of variables, to be determined for the model, is reduced. Thisalgorithm has been developed with GA, initially, and then with PSO and both implementations have similar performance on simulated EUTs with accurate results. Additionally in this chapter, using the aforementioned algorithm, four multi-magnetometer setups are studied in order to propose a more capable setup in obtaining a more “complete” magnetic signature of EUTs for the modeling.Finally, in chapter 8, the algorithms presented in chapters 4 and 7 are tested with real EUTs’ magnetic fields’ measurements and from the produced results their performance is, again, confirmed while, chapter 9 consists of the thesis conclusion and future work.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2016
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kakarakis, Sarantis-Dimitrios
- Κακαράκης, Σαράντης-Δημήτριος
Subjects
dc:subject × 20- Μαγνητική καθαρότητα
- Μοντελοποίηση με πολλαπλά δίπολα
- Αβεβαιότητα μετρήσεων
- Γενετικοί αλγόριθμοι
- Βελτιστοποίηση σμήνους μονάδων
- Μαγνητικό δίπολο
- Μαγνητικό τετράπολο
- Βαθμονόμηση
- Magnetic cleanliness
- Multiple dipole modeling
- Measuring uncertainty
- Genetic algorithms
- Particle swarm optimization
- Magnetic dipole
- Magnetic quadrupole
- Calibration
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/37631