kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Estimation and Forecasting of Load: A Comparison of a few Different Models
KTH, School of Electrical Engineering and Computer Science (EECS).
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

With the ambition to make society more sustainable, new obstacles have arisen in the electrical system. Theintroduction of electric vehicles and other low carbon technologies, together with weather-dependent renewableenergy generation, result in a more volatile power system. This poses challenges in operating the system and ensuringthat the electricity demand is consistently met. Estimating the future electric load is an important tool in optimizing theelectricity use. In this bachelor thesis, three different methods for day-ahead load forecasting are implemented on theSwedish bidding zone SE3 Stockholm. The selected methods are the ARIMA, Facebook Prophet and Random ForestRegression models. After evaluation, the results show that all models perform on a decent accuracy level compared to astate-of-the-art benchmark. Facebook Prophet and Random Forest Regression perform similarly and slightly better thanARIMA. The results of the two superior models show different strengths in terms of forecast accuracy over a year.Hence producing a combination of models is suggested for further research.

Abstract [sv]

I och med ambitionen att göra samhället mer hållbart har nya hinder uppkommit i elsystemet.Introduktionen av elbilar och annan utsläppssnål teknik, tillsammans med väderberoende förnybar energiproduktion,resulterar i ett mer volatilt elsystem. Detta medför utmaningar med driften av systemet och att försäkra att elbehovetalltid möts. Att förutspå den framtida elektriska lasten är ett viktigt verktyg för att optimera elanvändningen. I dettakandidatexamensarbete implementeras tre olika metoder för daglig lastprognostisering i det svenska elområdet SE3Stockholm. De valda metoderna är modellerna ARIMA, Facebook Prophet och Random Forest Regression. Efterutvärdering visar resultaten att alla modellerna presterar med en anständig precisionsnivå jämfört med bästatillgängliga referensram. Facebook Prophet och Random Forest Regression presterar på liknande nivå och något bättreän ARIMA. Resultatet av de två överlägsna modellerna visar olika styrkor i form av prognosriktighet över ett år. Därförföreslås att producera en kombination av modeller för framtida forskning.

Place, publisher, year, edition, pages
2024. , p. 295-304
Series
TRITA-EECS-EX ; 2024:158
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-358947OAI: oai:DiVA.org:kth-358947DiVA, id: diva2:1931063
Supervisors
Examiners
Projects
Kandidatexamensarbete Elektroteknik EECS 2024Available from: 2025-01-24 Created: 2025-01-24

Open Access in DiVA

fulltext(121150 kB)32 downloads
File information
File name FULLTEXT01.pdfFile size 121150 kBChecksum SHA-512
82342f1408fe2aae929f55e76f2a176a8521cf94c0fe100464225724a9b74ddff6f61a0fb1cf957b5d6400be30877b9d0aec28080253ae7c1e09793536e10217
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 32 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 612 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf