diplomsko delo
Tibor Čuš (Author), Tomaž Curk (Mentor), Maja Savinek (Co-mentor)

Abstract

V diplomskem delu analiziramo in modeliramo toplotne in elektriˇcne obre- menitve transformatorskih postaj s pomoˇcjo strojnega uˇcenja in numeriˇcnih metod. Transformatorske postaje so kljuˇcen element elektroenergetskega sis- tema, ki povezuje vire energije s konˇcnimi uporabniki. Zaradi vedno veˇcjega ˇstevila preobremenitev omenjenih postaj smo v diplomskem delu analizirali in modelirali njihove elektriˇcne in toplotne obremenitve. V ta namen so bile transformatorske postaje opremljene z dodatnimi temperaturnimi senzorji, ki so skupaj z vremenskimi podatki in podatki o porabi elektriˇcne energije tvorili naˇso podatkovno mnoˇzico. Na podatkih smo preizkusili veˇcje ˇstevilo modelov strojnega uˇcenja za napovedovanje odjema elektriˇcne energije. Naj- boljˇse rezultate so dosegli nakljuˇcni gozdovi in metoda podpornih vektorjev. Konˇcni rezultat diplomskega dela so napovedni modeli, ki se v kombinaciji z ekspertnim znanjem iz podroˇcja energetike lahko uporabljajo kot indikatorji preobremenitev elektroenergetskih transformatorjev.

Keywords

elektroenergetski sistem;napovedni modeli;indikatorji preobremenitev;visokošolski strokovni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [T. Čuš]
UDC: 004.8:621.314(043.2)
COBISS: 78691587 Link will open in a new window
Views: 266
Downloads: 57
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Modeling transformer station operation with machine learning methods
Secondary abstract: In this research, we analyze and model thermal and electrical energy loads of energy transformer stations with the help of machine learning and numerical methods. Transformer stations are a key part of the electrical power system. They are the elements that connect energy sources to end-users. Because of an ever-increasing amount of transformer station overloads, this thesis focuses on analyzing and modeling thermal and electrical loads. For this reason, transformer stations have been equipped with temperature sensors. We combined transformer station temperature data with weather and energy usage data. We used multiple machine learning algorithms to predict elec- trical energy consumption. The best results were obtained by random forest and support vector machines. Our research results are forecasting models that can be combined with expert domain knowledge to predict transformer station overloads.
Secondary keywords: electrical power system;transformer station;machine learning;forecasting models;overload indicators;computer science;diploma;Transformatorske postaje;Strojno učenje;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000470
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 55 str.
ID: 14306021