izpis_h1_title_alt

Pregled in primerjava metahevrističnih pristopov pri določanju dopustnih vrednosti parametrov dinamičnega sistema
ID GRDADOLNIK, ALJAŽ (Author), ID Moškon, Miha (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (3,47 MB)
MD5: 594B03A7DD0B51F81F0AF655A6034FB4

Abstract
V nalogi se osredotočimo na model dinamičnega sistema, ki ga lahko opišemo s sistemom diferencialnih enačb. V teh nastopajo številni parametri, katerih točnih vrednosti ne poznamo oziroma jih težko določimo. Vseeno pa nas zanima, kako dobro podan sistem enačb opisuje opazovan sistem in pri kakšnem razponu vrednosti parametrov sistem odraža želeno dinamiko. Reševanja tega problema se lotimo z različnimi metahevristikami za globalno optimizacijo. V našem primeru so to algoritmi optimizacija sivega volka (angl. \emph{Grey-Wolf Optimization}, GWO), optimizacija kita (angl. \emph{Whale Optimization Algorithm}, WOA) ter optimizacija z rojem delcev (angl. \emph{Particle Swarm Optimization}, PSO). Z njimi poizkušamo čim bolje prečesati prostor dopustnih vrednosti parametrov in določiti točke oziroma regije, kjer sistem deluje kot bi moral.

Language:Slovenian
Keywords:algoritem sivi volk, algoritem optimizacija kita, optimizacija z rojem delcev, analiza robustnosti, preiskovanje prostora, superračunalnik
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FRI - Faculty of Computer and Information Science
Year:2022
PID:20.500.12556/RUL-140425 This link opens in a new window
COBISS.SI-ID:123601155 This link opens in a new window
Publication date in RUL:14.09.2022
Views:693
Downloads:112
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Overview and comparison of metaheuristics for assessment of feasible parameter regions in a dynamical system
Abstract:
As part of this thesis we have a model of a random dynamic system, which we can describe with a system of diferential equation. In it there are multiple parameters of which the exact values we do not know or they are difficult to determine. Still we are interested in how well the system of equations describes the observed system, and over what range of parameter values does the system reflect the desired dynamics. We try to solve this problem with different metaheuristics for global op- timization. In our case this are the grey-wolf optimization, the whale opti- mization algorithm and particle swarm optimization. With them we try to search a space of acceptable parameter values and try to identify the points of regions where the system is working correctly.

Keywords:Grey wolf optimizer, Whale optimization algorithm, Particle swarm optimization, robustness analysis, exploring search space, supercomputer

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back