doctoral thesis
Fault diagnosis in distribution network using fuzzy logic system

Slaven Kaluđer (2014)
Sveučilište Josipa Jurja Strossmayera u Osijeku
Fakultet elektrotehnike, računarstva i informacijskih tehnologija Osijek
Zavod za elektroenergetiku
Katedra za elektroenergetske mreže i postrojenja
Metadata
TitleDIJAGNOZA KVARA U DISTRIBUTIVNOJ MREŢI POMOĆU SUSTAVA S NEIZRAZITOM LOGIKOM
AuthorSlaven Kaluđer
Mentor(s)Lajos Jozsa
Abstract
U doktorskoj disertaciji razvijen je neizraziti ekspertni sustav (off-line) za dijagnozu kvara u distributivnoj mreţi na bazi strukturalnog i funkcionalnog djelovanja releja i prekidača. Prikazan je novi način modeliranja funkcionalnog djelovanja releja i prekidača koji je opisan pomoću neizrazite logike. Neizrazita logika (više-vrijednosna logika) je metoda koja se bavi modeliranjem nepreciznosti, nesigurnosti kod donošenja odluka. Poboljšava i povećava potencijal kod modeliranja ljudskog razmišljanja koje koristi lingvističke varijable. Neizrazita logika (teorija) je u suprotnosti s klasičnom teorijom vjerojatnosti, jer je prva deterministička, a druga stohastička u naravi. Teorija vjerojatnosti se koristi bi-vrijednosnom logikom (0 ili 1), dok neizrazita teorija moţe poprimiti bilo koju vrijednost izmeĎu 0 i 1. Neizrazita logika nudi jednostavan pristup modeliranju nejasnih, neprecizni, izobličeni i nepostojećih ulaznih informacija. Neizraziti dijagnostički sustav kao ulazne podatke koristi statuse releja i prekidača te njihove vremenske oznake. Dijagnostički sustav iz ogromnog broja alarma postavlja, logički organizira i kvantificira dijagnoze. Razvijeni dijagnostički sustav za dijagnozu kvara uspješno dijagnosticira djelovanje, krivo djelovanje i otkazivanje djelovanja releja i prekidača. TakoĎer ima mogućnost lociranja (identificiranja) kvara te njegovog kvantificiranja na bazi hibridnog operatora unije koji je iskazan kao aritmetička sredina Hamackerovog i Schweizerovog operatora unije. Pored navedenog uspješno se nosi i s kvarovima telemetrije koristeći novi način modeliranja nepostojećih informacija pomoću linearne funkcije pripadnost u odnosu na vrijeme.
Keywordsfault diagnosis alarm processing fuzzy logic fuzzy expert systems.
Parallel title (English)Fault diagnosis in distribution network using fuzzy logic system
Committee MembersSrete Nikolovski (committee chairperson)
Predrag Marić (committee member)
Marinko Barukčić (committee member)
Ivica Pavić (committee member)
GranterSveučilište Josipa Jurja Strossmayera u Osijeku
Fakultet elektrotehnike, računarstva i informacijskih tehnologija Osijek
Lower level organizational unitsZavod za elektroenergetiku
Katedra za elektroenergetske mreže i postrojenja
PlaceOsijek
StateCroatia
Scientific field, discipline, subdisciplineTECHNICAL SCIENCES
Electrical Engineering
Power Engineering
UDK621.3
APPLIED SCIENCES. MEDICINE. TECHNOLOGY
Electrical engineering
Study programme typeuniversity
Study levelpostgraduate
Study programmePostgraduate doctoral study programme; branch Power Engineering and Communications, branch Informatics
Study specializationbranch Power Engineering and Communications
Academic title abbreviationdr.sc.
Genredoctoral thesis
Language Croatian
Defense date2014-11-18
Promoted date2016-04-09
Parallel abstract (English)
In this doctoral dissertation was developed off-line fuzzy expert system for fault diagnosis in distribution network based on the structural and functional operation of relays and circuit breakers. Functional operation of relays and circuit breakers is described by fuzzy logic as a new way of modeling. Fuzzy logic (multiple-value logic) is a method that deals with the modeling of imprecision, uncertainty in decision-making process. It improves and increases the potential in the modeling of human thinking which uses linguistic variables. Fuzzy logic (theory) is in contrast with the classical theory of probability, since the first is deterministic and the other is stochastic in nature. Probability theory uses bi-value logic (0 or 1), while fuzzy theory can take any value between 0 and 1. Fuzzy logic offers a simple approach for modeling vague, inaccurate, distorted and missing informations. Imput information of fuzzy diagnostic system are the status of relays and circuit breakers and their time stamps. Diagnostic system logically organize and quantifie diagnosis from the huge alarm streams.Operation, false operation and failure of operation for circuit brakers and realays can be diagnosed with the developed diagnostic system. Fault location (identification) is quantified with hybrid union operator which is discribed as the arithmetic mean of Hamackers and Schweizers operator. New method of modeling the missng information as linear membership functions with respect to time is used for dealing with failures of telemetry.
Parallel keywords (Croatian)dijagnoza kvara procesiranje alarma neizrazita logika neizraziti ekspertni sustavi.
Extent160
Versionaccepted version
Resource typetext
Access conditionOpen access
Terms of usehttp://rightsstatements.org/vocab/InC/1.0/
Noteaccepted version
URN:NBNhttps://urn.nsk.hr/urn:nbn:hr:200:904233
CommitterAnka Ovničević