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
Short-Term Heat Load Forecasting in District Heating Systems: A Comparative Study of Various Forecasting Methods
KTH, School of Electrical Engineering and Computer Science (EECS).
2019 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Short term heat load forecasts are vital for optimal production planning and commitmentof generation units. The generation utility also bares balance responsibilitytoward the electricity market as a result of CHP generation. Sub-optimal loadforecasts can lead to high costs relating to unit commitment, fuel usage and balancingcosts. This thesis presents the empirical comparison of various models for 24hheat load forecasting. Five methods were investigated including four supervisedmachine learning algorithms; neural networks, support vector machines, randomforests and boosted decision trees and one autoregressive time series model; ARIMAX.The models were developed, and evaluated using cross validation with oneyear of heat load data from a local district heating system and corresponding meteorologicaldata from the same time period. The thesis also investigates the impactof feature selection on predictive power and generalization ability. The results indicatea significant difference in forecast accuracy between the methods with neuralnetworks and ARIMAX showing the best and similar performance followed by thesupport vector machine, boosted decision trees and random forest.

Abstract [sv]

Korttidsprongoser för fjärrvärmelast är mycket viktiga för optimal produktionsplanering.Energibolag som använder kraftvärme bär dessutom balansansvar gentemotelmarknaden. Sub-optimala lastprognoser kan leda till höga kostnader för start ochstopp, bränsleåtgång och obalanser. Detta examensarbete presenterar den empiriskajämförelsen av olika metoder avseende 24-timmars lastprognostisering. Totalt femmetoder undersöktes varav fyra maskininlärningsalgoritmer; neurala nätverk, stödvektormaskin,random forest samt boosted desicion trees och en tidsseriemodell;ARIMAX. Modellerna utvecklades, och utvärderades med hjälp av korsvalideringpå ett års lastdata från ett lokalt fjärrvärmenät och motsvarande väderdata för sammatidsperiod. Examensarbetet undersöker även inverkan av variabelselektion på prognosernasprecision och förmåga att generalisera. Resultaten tyder på en signifikantskillnad i noggrannhet mellan de olika modellerna. Bäst resultat uppnåddes av neuralanätverk och ARIMAX med en liten skillnad sinsemellan, följt av stödvektormaskin,boosted decision trees och random forest.

Place, publisher, year, edition, pages
2019. , p. 76
Series
TRITA-EECS-EX ; 2019:573
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:kth:diva-265670OAI: oai:DiVA.org:kth-265670DiVA, id: diva2:1380775
Educational program
Master of Science in Engineering - Electrical Engineering
Supervisors
Examiners
Available from: 2019-12-19 Created: 2019-12-19 Last updated: 2019-12-19Bibliographically approved

Open Access in DiVA

fulltext(3315 kB)4 downloads
File information
File name FULLTEXT01.pdfFile size 3315 kBChecksum SHA-512
67657cd94f4aaefef5820c7bcc4f126e7107e6dcc1009e906e51fc56df1ef5a1b8f2d3387e48e6ce11a6c4d23ce599b561cd4a2496380ba77c4cd2dc3ceed2a8
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 4 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: 24 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