Kratkoročno predviđanje opterećenja električne mreže pomoću metode potpornih vektora

scientific master's theses

• Title (english): Short term electric load forecasting based on support vector machines
• Author: Ervin Čeperić
• Mentor: Damir Kalpić (mentor)
• Committee member: Davor Škrlec (predsjednik povjerenstva)
• Committee member: Damir Kalpić (član povjerenstva)
• Committee member: Robert Manger (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Applied Computing); Zagreb
• Defense date and country: 2011-01-25, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Planiranje opterećenja u električnoj mreži jedna je od važnih komponenata u planiranju i radu elektroenergetske industrije. Precizna procjena opterećenja presudna je u predviđanju cijena električne energije te može imati veliki utjecaj na prijenos električne energije, tržišne udjele i dobiti u konkurentnim elektroenergetskim tržištima. Ovaj rad usredotočuje se na kratkoročno planiranje opterećenja u električnoj mreži. Razvijena je metodologija kratkoročnog planiranja temeljena na metodi potpornih vektora. Kao ključan čimbenik uspjeha planiranja, identificirana je prilagodba modela za pojedinu regiju ili skup podataka, temeljena na prethodnoj analizi opterećenja i raznih specifičnosti regije ili skupa podataka. U ovom radu upravo je takva prilagodba izvršena na dva primjera, elektroenergetskog sustava Slovačke (na temelju EUNITE natjecanja) i trafostanice 110/35 kV Lošinj. Analiza rezultata predviđanja pokazuje korisnost modela i perspektivu za upotrebu u elektroenergetskoj industriji.

Abstract (english)

Load forecasting has become one of the most crucial parts in planning and operations of electric power industry. An accurate estimation of load is essential for electricity price forecasting and can have significant implications to energy transactions, market shares and profits in competitive electricity markets. This paper focuses on short-term load forecasting (STLF). The methodology for STLF is presented, based on support vector machines. Adaptation to regional specificities, based on analysis of its electric load, weather and economy, is identified as the key factor for accurate forecasting. The forecasting methodology, model and adaptation strategy is tested on two examples, Slovak electric power system (based on EUNITE competition) and electric power subsystem 110/35 kV Lošinj. The analysis of prediction results shows that the developed models have perspective for use in electric power industry.

• Keywords: planiranje opterećenja u električnoj mreži; kratkoročno planiranje opterećenja u električnoj mreži; metoda potpornih vektora; algoritam za optimizaciju temeljen na imitaciji ponašanja rojeva
• Keywords (english): load forecasting; short-term load forecasting; support vector machines; particle swarm optimization
• Language: croatian
• URN:NBN: urn:nbn:hr:168:995711
• Promotion: 2020
• Project: Number: 036-0361983-2022; Title: Održivi razvoj informacijskih sustava; Leader: Krešimir Fertalj; Jurisdiction: Croatia; Funder: MZOS; Funding stream: ZP
• Project: Number: 036-0361983-3137; Title: Optimiranje i upravljanje rizicima u informacijskim sustavima; Leader: Damir Kalpić; Jurisdiction: Croatia; Funder: MZOS; Funding stream: ZP
• Study programme: Title: Postgraduate master programe in computing; Study programme type: university; Study level: postgraduate; Academic / professional title: magistar inženjer računarstva (magistar inženjer računarstva)
• Catalog URL: http://lib.fer.hr/cgi-bin/koha/opac-detail.pl?biblionumber=34377
• Type of resource: Text
• Extent: 69 str ; 30 cm
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-08-21 08:42:43