Download
PDF 2.35 MB
master's thesis
28   12
Electricity price forecasting
Ines Klešić (2018)
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Power Engineering
Chair of Power Systems and Substations
Metadata
TitlePrognoza cijene električne energije
AuthorInes Klešić
Mentor(s)Srete Nikolovski (thesis advisor)
Krešimir Fekete (thesis advisor)
Abstract
Diplomski rad podijeljen je na teorijski i praktični dio. Ovisno o vrsti primjene, predviđanje opterećenja i cijena dijeli se u tri glavne kategorije: kratkoročno, srednjoročno i dugoročno. Faktori koji utječu na obrasce opterećenja i cijene su opisani. ANN metoda se koristi jer pruža jednostavan i učinkovit način za prognozu u praktičnoj primjeni te također predstavlja i nelinearne funkcije. Načini učenja i metode treniranja su pojašnjeni. Praktični dio najprije opisuje podatke za trening i testiranje koji su korišteni za predloženi model za kratkoročnu prognozu opterećenja i cijene. Opterećenje, cijena i temperatura su prikazani. Rezultati su razmotreni na temelju srednje apsolutne postotne pogreške.
Keywordsartificial neural networks short-term load forecast short-term price forecast mean absolute percentage error volatility
Parallel title (English)Electricity price forecasting
Committee MembersSrete Nikolovski (committee chairperson)
Goran Knežević (committee member)
Krešimir Fekete (committee member)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Lower level organizational unitsDepartment of Power Engineering
Chair of Power Systems and Substations
PlaceOsijek
StateCroatia
Scientific field, discipline, subdisciplineTECHNICAL SCIENCES
Electrical Engineering
Power Engineering
Study programme typeuniversity
Study levelgraduate
Study programmeGraduate University Study Programme in Electrical Engineering; specializations in: Communications and Informatics, Power Engineering
Study specializationPower Engineering
Academic title abbreviationmag.ing.el.
Genremaster's thesis
Language Croatian
Defense date2018-03-14
Parallel abstract (English)
The final thesis has theoretical and the pratical part. Based on its specific application, load and price forecasting can be categorized into short-term, mid-term and longterm. Factors affecting load and price patterns are described. ANN method is used because it provides a simple and powerful tool for forecasting in practical systems and it also represents non-linear functions. Learning strategies and training methods are explained. Practical section first described training and test data used to test the proposed ANN model for short-term load and price forecasting. Load, price and temperature data are presented. Then results are discussed based on the mean absolute percentage error.
Parallel keywords (Croatian)umjetne neuronske mreže kratkoročna prognoza opterećenja kratkoročna prognoza cijene srednja apsolutna postotna pogreška volatilnost
Resource typetext
Access conditionOpen access
Terms of usehttp://rightsstatements.org/vocab/InC/1.0/
URN:NBNhttps://urn.nsk.hr/urn:nbn:hr:200:003882