- author = "Kateryna Stoyka",
- title = "Methods and Algorithms for Behavioral Modeling of Ferrite Power Inductors",
- school = "Dipartimento Ingegneria dell'Informazione ed Elettrica e Matematica Applicata, Universita degli studi di Salerno",
- year = "2019",
- address = "Italy",
- month = "12 " # mar,
- keywords = "genetic algorithms, genetic programming, Inductor Power Loss Model, FPI, SMPS, Sustainable Saturation Operation",
-
URL = "
http://elea.unisa.it:8080/xmlui/handle/10556/4513",
-
URL = "
http://elea.unisa.it/jspui/bitstream/10556/4513/1/tesi%20di%20dottorato%20K.%20Stoyka.pdf",
-
abs_url_en = "http://elea.unisa.it/jspui/bitstream/10556/4513/2/abstract%20in%20inglese%20K.%20Stoyka.pdf",
-
abs_url_it = "http://elea.unisa.it/jspui/bitstream/10556/4513/3/abstract%20in%20italiano%20K.%20Stoyka.pdf",
-
DOI = "
doi:10.14273/unisa-2711",
- size = "221 pages",
-
abstract = "Information technology allows solving numerous
problems regarding all the aspects of everyday life,
including technical activities related to the design of
devices and systems. In the electronic field, different
types of software are widely used to support designers
in solving the problems of electronic circuit design,
at device level and system level. Power electronics is
one of the most important modern technologies, since
power supply systems are used to feed any electric and
electronic device and system in manifold applications
(e.g. computers, automotive, aerospace, consumer
electronics, etc). Switching power supply design is
mostly driven by high efficiency and high reliability
requirements. The strong non linearity of switching
power supplies and the difficulty of application of
advanced design methodologies often push designers to
adopt a conservative approach, based on simplified
robust and reliable methods. This mostly result in
sub-optimal design solutions characterised by
components over sizing. This dissertation discusses
innovative applications of enhanced numerical
techniques and intelligent algorithms to power supplies
optimisation and design. The impact of innovative
modeling and computing techniques in the discovery of
novel advanced solutions outperforming the traditional
conservative designs is emphasised.
Power electronics is ever moving towards higher efficiency and higher power density. Magnetic components inductors and transformers occupy a significant amount of space in today Switch-Mode Power Supplies (SMPSs), and furthermore, considerable losses occur in these components. In order to achieve a higher level of miniaturization, reduction in the size of these components is crucial. Ferrite Power Inductors (FPIs) are usually the first choice for high-efficiency designs of SMPS, thanks to their resulting low losses. However, FPIs suffer of a pretty sharp inductance drop when their current exceeds a certain threshold, occurring due to the saturation of their magnetic 2 core. In SMPS design, it is commonly considered a good practice to select FPIs operating in the region of weak saturation (within about 20 percent inductance drop). This limitation is due to the lack of methods for quick prediction of real impact of FPIs saturation in SPMS applications. The consequence of the adoption of such conventional design approach is that inductors are often oversized....
a) development of numerical techniques and intelligent algorithms for generation and discovery of behavioral models for saturation and power losses of FPIs used in SMPS applications
b) development of enhanced numerical algorithms, using the above models, able to reliably predict the FPIs behavior under given SMPS conditions
c) development of enhanced numerical algorithms able to identify feasible inductor solutions, possibly operating in saturation, allowing to reduce the inductor size and increase the converter power density.
...",
-
notes = "
Supervisor: Nicola Femia, Co-Supervisor: Giulia Di Capua",
