Primjena metoda umjetne inteligencije pri izboru materijala

doctoral thesis

• Title (english): Application of artifical intelligence methods in materials selection
• Author: Dragutin Lisjak
• Mentor: Tomislav Filetin (mentor)
• Committee member: Branko Novaković (predsjednik povjerenstva)
• Committee member: Tomislav Filetin (član povjerenstva)
• Committee member: Mladen Stupnišek (član povjerenstva)
• Committee member: Bojan Jerbić (član povjerenstva)
• Committee member: Darko Grundler (član povjerenstva); MBZ: 173300
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2004-11-24, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering
• Universal decimal classification (UDC): 620 - Materials testing. Power stations. Economics of energy ; 004 - Computer science and technology. Computing. Data processing

Abstract

U radu je sažeto dan prikaz aktualnih metoda umjetne inteligencije: neuronskih mreža, neizrazite
logike, genetičkih algoritama i ekspertnih sustava. Detaljno je opisana struktura zamišljenog
inteligentnog sustava za izbor materijala u uvjetima trošenja kao i način za poboljšanje njegove
glavne komponente mehanizma zaključivanja. Također su opisani i podmehanizmi glavnog
mehanizma zaključivanja: mehanizam za obradu konverzacije, mehanizam za obradu asocijacija,
mehanizam za obradu pravila mehanizam za obradu kriterija i izbor materijala, mehanizam za
izbora postupaka modificiranja površina. U eksperimentalnom dijelu rada dan je plan izgradnje
inteligentnog sustav za izbor materijala, način prikaza rezultata eksperimenta te pregled
softverskih alata. Prikazani su načini mapiranja ulazno-izlaznih skupova za učenje neuronskih
mreža i mehanizama zaključivanja neizrazite (fuzzy) logike. Prikazan je način kreiranja
vremenskih asocijativnih matrica odgovornih za dobivanje i objašnjenje rješenja razmatranog
problema. Prikazan je način optimiranja parametara neuronskih mreža genetičkim algoritmima,
kao i primjena NeuroFuzzy modela neizrazite logike. Dan je konkretni primjer načina rada
mehanizma zaključivanja inteligentnog sustava za izbor materijala.

Abstract (english)

In the work, current artifical intelligence methods are concisely presented: neural networks,
fuzzy logic, genetic algorithms and expert systems. Structure of the designed intelligent system
for material selection in tribological conditions is described detailedly, as well as the ways for
improvement of its main inference engine component. The subengines of the main inference
engine are also presented: conversation processing engine, associations processing engine, rules
processing engine, criteria processing & material selection engine, engine for selection of surface
modification processes. In experimental part of the work, concept for construction of the
intelligent system for material selection is given, as well as the method how to present the
experimental results and an overview of software tools. Mapping methods of the input-output
data for training of neural networks and inference engines using fuzzy logic are presented.
Procedure for creating of the time associative matrices which manage obtaining and explanation
of the solution of the analyzed problem is presented. Optimizing of neural network using genetic
algorithms is presented, as well as application of NeuroFuzzy model of fuzzy logic. An actual
example illustrating operating principle of inference engine of the intelligent system for material
selection is provided.

• Keywords: izbor materijala; umjetna inteligencija; neuronske mreže; neizrazita logika; genetički algoritmi; ekspertni sustavi
• Keywords (english): material selection; artificial intelligence; neural networks; fuzzy logic; genetic algorithms; expert systems
• Language: croatian
• URN:NBN: urn:nbn:hr:235:624569
• Study programme: Title: Mechanical Engineering and Naval Architecture; Study programme type: university; Study level: postgraduate; Academic / professional title: doktor/doktorica znanosti, područje tehničkih znanosti (doktor/doktorica znanosti, područje tehničkih znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2020-12-04 13:54:50