Abstract
In the recent years machine learning algorithms have developed further and various applications are taking advantage of this advancement. Modern machine learning is now used in district heating for more precise and realistic heat demand prediction. Machine learning methods like Artificial Neural Network (ANN), Linear Regression (LR), and Decision Tree (DT) are commonly adopted in heat demand prediction to produce more accurate results. This research paper compares the performance of several machine learning methods on different datasets generated by the combination of simulations and real-life data collected from a local district heating site in Nottingham. The result shows that Linear Regression generates better prediction than Artificial Neural Network and Decision Tree, for dataset generated using simulator, whereas Decision Tree performs best for real-life data.
Highlights
District heating (DH) system plays vital role to achieve low carbon emission by 2050
The use of statistical method as a kind of machine learning model for prediction of loads in district-heating systems is first described in the Simple model based on social behavior and outdoor temperature by Dotzauer [4]
The advantage of having a greater number of features can be leveraged by using machine learning algorithms like Decision Tree and Artificial Neural Network
Summary
District heating (DH) system plays vital role to achieve low carbon emission by 2050. The use of statistical method as a kind of machine learning model for prediction of loads in district-heating systems is first described in the Simple model based on social behavior and outdoor temperature by Dotzauer [4]. Many researchers started to explore heat prediction for district heating using time-series with the inclusion of outdoor temperature [5]. Various researchers started using the machine learning algorithms in District heating for heat demand prediction [6,7]. Talebi et al [8] evaluates various heat prediction techniques used in the past. It describes the complexity level by four parameters viz. The paper classifies district heating based on geographical conditions, scale, heat density and end-user demand. Support vector machine (SVM) is used where small datasets are available
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
