hh.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
Detecting Faults in Customer Substations to Reduce the High Return Temperature in District Heating
Halmstad University, School of Information Technology.
Halmstad University, School of Information Technology.
2025 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This study aimed to develop different methods of artificial intelligence to detect faults in district heating substations to reduce high return temperatures, in collaboration with Halmstad Energi och Miljö (HEM). Three approaches were applied: the statistical model Autoregressive Integrated Moving Average (ARIMA), the autoencoder-based approach using UnSupervised Anomaly Detection model (USAD), and the machine learning k-Nearest Neighbors algorithm (k-NN). Using unlabeled data from more than 4,000 buildings (2022–2025), faults were injected to evaluate the performance of the models. ARIMA used time series forecasting, achieving 51.1 % recall and 16.9 % precision. USAD demonstrated balanced performance, with 36.9 % recall and 36.8 % precision, while k-NN showed high recall, 77.4 %, and precision, 35.3 %. The study shows the precision-recall tradeoff, with method selection depending on how powerful the model is at detecting faults. Although k-NN shows high performance in detecting injected faults, minimizing the number of false alarms is required for operational deployment.

Abstract [sv]

Syftet med denna studie är att utveckla olika metoder av artificiell intelligensför att upptäcka fel i fjärrvärmeundercentraler med avsikt att minska höga re-turtemperaturer, i samarbete med Halmstad Energi och Miljö (HEM). Tre metodertillämpades: den statistiska modellen Autoregressive Integrated Moving Average(ARIMA), en autoencoder-baserad metod med modellen UnSupervised AnomalyDetection (USAD), samt maskininlärningsalgoritmen k-Nearest Neighbors (k-NN).Med hjälp av oetiketterad data från över 4 000 byggnader (2022–2025) injicer-ades fel för att utvärdera modellernas prestanda. ARIMA använde tidsserieprog-noser i kombination med tröskelvärden baserade på statistisk processtyrning (SPC),vilket resulterade i 51,1% recall och 16,9% precision. USAD visade en mer bal-anserad prestanda med 36,9% recall och 36,8% precision, medan k-NN uppnåddehög recall 77,4% och precision 35,3%. Studien belyser avvägningen mellan preci-sion och recall, där val av metod beror på hur effektiv modellen är på att upptäckafel. Även om k-NN visar hög träffsäkerhet i att identifiera injicerade fel, krävs detatt antalet falska larm minimeras för att modellen ska kunna användas i praktiskdrift.

Place, publisher, year, edition, pages
2025.
Keywords [en]
District heating systems, fault detection, anomaly detection, ARIMA, USAD, k-NN, energy efficiency, unsupervised anomaly detection, high return tem- perature, static, machine learning, deep learning
Keywords [sv]
Fjärrvärme, feldetektering, avvikelsedetektering, ARIMA, USAD, k- NN, energieffektivitet, oövervakad avvikelsedetektering, hög returtemperatur, statisk, maskininlärning, djupinlärning.
National Category
Artificial Intelligence Computer Sciences Computer Engineering
Identifiers
URN: urn:nbn:se:hh:diva-56221OAI: oai:DiVA.org:hh-56221DiVA, id: diva2:1964764
External cooperation
Halmstads Energi & Miljö
Supervisors
Examiners
Available from: 2025-06-05 Created: 2025-06-05 Last updated: 2025-10-01Bibliographically approved

Open Access in DiVA

fulltext(1405 kB)224 downloads
File information
File name FULLTEXT02.pdfFile size 1405 kBChecksum SHA-512
ad35497a603200c90aec4f8aeb1864c90aa641568bd4f4e91e066a4357ef79128133d7739d4149f032eb7640d0c4951da32d9a78af898b35a201c509b017c404
Type fulltextMimetype application/pdf

By organisation
School of Information Technology
Artificial IntelligenceComputer SciencesComputer Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 224 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: 276 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