Sustavi promjenjive strukture za vođenje plovila po trajektoriji

scientific master's theses

• Author: Edin Omerdić
• Mentor: Zoran Vukić (mentor)
• Committee member: Zoran Vukić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Control and Computer Engineering); Zagreb
• Defense date and country: 2001, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Process Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

U radu je dan pregled metoda za dijagnostiku kvara i sintezu rekonfigurabilnog sustava upravljanja. Također su dane osnove samo-organizirajućih mapa, koje mogu poslužiti kao pouzdan klasifikator kvarova. Prikazana je arhitektura i tehnike učenja adaptivnih mreža. Prikazana je klasa adaptivnih mreža ANFIS, koja je funkcionalno ekvivalentna neizrazitom sustavu zaključivanja. Opisana je paradigma zvana neuro-neizrazito upravljanje. Detaljno su opisane sljedeće metode sinteze: inverzno učenje, specijalizirano učenje, sinteza samo-organizirajućeg regulatora s povratnim prostiranjem kroz vrijeme i sinteza samo-organizirajućeg regulatora sa "meta" pravilima.
U radu je prikazana sinteza sustava upravljanja promjenjive strukture, otpornog na kvarove, na primjeru vođenja broda Mariner klase po kursu i trajektoriji. Prikazana je metodologija za razvoj sustava vođenja broda po kursu i trajektoriji, koji je otporan na kvarove (kvar u izvršnom organu - aktuatoru, kvar na senzoru kursa - žirokompasu i kvar na senzoru pozicije broda - GPS) uz prisustvo poremećaja (valovi i morske struje).

Abstract (english)

An overview of the methods for fault diagnosis and design of reconfigurable control system is given. Basics of self-organizing maps, which can be used as reliable fault classificator, are given in detail. Architecture and learning procedures of adaptive networks are described. ANFIS, a special class of adaptive network, which is functionally equivalent to fuzzy inference system, is described. Paradigm neuro-fuzzy control is described. Different design methods (inverse learning, specialised learning, self-organizing controller with back-propagation through time and self-organizing controller with "meta" rules) are given in detail.
Design of restructurable fault tolerant control system for ship course-keeping and track-keeping is given (Mariner class vessel). Methodology for development of control system for ship course-keeping and track-keeping, which is robust on faults in actuator, gyrocompass and GPS in presence of disturbances (waves and sea currents), is described.

• Keywords: Adaptivne mreže; ANFIS; detekcija i izolacija kvara; dijagnostika kvara; neuro-neizrazito upravljanje; samo-organizirajuće mape; sustav upravljanja otporan na kvarove; rekonfigurabilni sustav upravljanja; vođenje broda po kursu i trajektoriji
• Keywords (english): Adaptive Networks; ANFIS; Fault Detection and Isolation; Fault Diagnosis; Fault Tolerant Control System; Neuro-Fuzzy Control; Self-Organizing Maps; Ship Course-Keeping and Track-Keeping; Reconfigurable Control System
• Language: croatian
• URN:NBN: urn:nbn:hr:168:112703
• Study programme: Title: Postgraduate master programe in computing; Study programme type: university; Study level: postgraduate; Academic / professional title: magistar inženjer računarstva (magistar inženjer računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-04-15 13:41:55