Usporedba rada evolucijskog algoritma i algoritma roja na standardnim testnim funkcijama

undergraduate thesis

• Title (english): Comparison of evolutionary and swarm algorithm on standard benchmark test functions
• Author: Lovro Čehulić
• Mentor: Petar Ćurković (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2017-03-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Genetski algoritam i optimizacija rojem čestica su metaheuristički optimizacijski alati. Genetski algoritam pripada području evolucijskog računarstva, a inspiriran je evolucijom na genetskoj razini. Optimizacija rojem čestica pripada području inteligencije roja i inspirirana je rojevima kukaca i jatima ptica čija je specijalizacija problem traženja. U ovome radu, ova dva optimizacijska algoritma testirana su na raznim testnim funkcijama te su dobivena dobra rješenja uz određena ograničenja. Odabrane testne funkcije su uglavnom multimodalni, višedimenzijski minimizacijski problemi. Optimizacijom rojem čestica postignuti su rezultati visoke točnosti za probleme koji su uspješno optimizirani. Genetski algoritam dao je rezultate koji nisu tako visoke točnosti, ali su njime pronađena rješenja onih funkcija za koje Optimizacija rojem čestica nije dala rezultate.

Abstract (english)

Genetic Algorithm and Particle Swarm Optimization are Metaheuristic optimization tools. Genetic Algorithm belongs to the field of evolutionary computing, and it is inspired by evolution on genetic level. Particle Swarm Optimization belong to the field of swarm inteligence inspired by swarms of insects and flocks of birds whose specialty is search problem. In this thesis, these two optimization algorithms were tested on various (benchmark) test functions and have obtained good solutions with certain restrictions. The selected test functions are mainly multimodal, multidimensional minimization problems. Particle Swarm

Optimization results have been achieved with high precision on the successfully optimize d problems. Genetic Algorithm has provided results with not as high accuracy, but it found solutions for those functions that Particle Swarm Optimization did not.

• Keywords: Genetski algoritam; optimizacija rojem čestica; standardne testne funkcije.
• Keywords (english): Genetic Algorithm; Particle Swarm Optimization; standard (benchmark) test functions.
• Language: croatian
• URN:NBN: urn:nbn:hr:235:633835
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2019-05-22 19:04:41