A novel machine learning approach to predict shortterm energy load for future low-temperature district heating

[featured_image]
Download
Download is available until [expire_date]
  • Version
  • Download 100
  • File Size 1.35 MB
  • File Count 1
  • Create Date September 12, 2022
  • Last Updated November 14, 2022

A novel machine learning approach to predict shortterm energy load for future low-temperature district heating

Conference paper presented in Clima 2022 held in Rotterdam, 22-25 May

Title: A novel machine learning approach to predict shortterm energy load for future low-temperature district heating

Language: English

Authors: Thomas Ohlson Timoudas (*a), Yiyu Ding (*b), Qian Wang (*c,d)

*a: RISE Research Institutes of Sweden, Sweden, thomas.ohlson.timoudas@ri.se

*b: Department of Energy and Process Engineering, Norwegian University of Science and Technology (NTNU),
Trondheim, Norway, yiyu.ding@ntnu.no

*c: Department of Civil and Architectural Engineering, KTH Royal Institute of Technology, Brinellvägen 23, Stockholm,
Sweden, qianwang@kth.se

*d: Uponor AB, Hackstavägen 1, Västerås, 721 32, Sweden

Abstract: In this work, we develop machine learning methods to forecast the day-ahead heating energy demand of district heating (DH) end-users in hourly resolution, using existing metering data for DH end-users and weather data. The focus of the study is a detailed analysis of the accuracy levels of short-term load prediction methods. In particular, accuracy levels are quantified for  Artificial Neural Network (ANN) models with variations in the input parameters. The importance of historical data is investigated – in particular the importance of including historical hourly heating loads as input to the forecasting model. Additionally, the impact of different lengths of the historical input data is studied. Our methods are evaluated and validated using metering data from a live use-case in a Scandinavian environment, collected from 20 DH-supplied nursing homes through the years of 2016 to 2019. This study demonstrates that, although there is a strong linear relationship between outdoor temperature and heating load, it is still important to include historical heating loads as an input for prediction of future heating loads. Furthermore, the results show that it is important to include historical data from at least the preceding 24 hours, but suggest diminishing returns of including data much further back than that. The resulting models demonstrate the practical feasibility of such prediction models in a live use-case.