Scholarlink

This paper shows the application of supervised machine learning methods for predicting the gas consumption of the district heating system at the Faculty of Mechanical Engineering in Nis. The district heating system is automatically controlled, while its energy efficiency is continuously monitored through the SCADA system. Although the SCADA system provides detailed operational data, optimization decisions aimed at reducing energy consumption and costs are made by the heating plant operator. The focus of this research is on predicting natural gas consumption using the decision tree method, a machine learning technique, applied to a dataset encompassing the heating season from October 15, 2022, to December 31, 2022. The dataset includes various energy indicators obtained from the SCADA system of the heating plant, such as weather conditions, historical consumption, and other system parameters. MATLAB software was utilized for the development and testing of the model. The model's predictions were validated using data from October 15, 2024, to December 31, 2024 to evaluate its performance on unseen data. The primary objective of this study is to demonstrate the effectiveness of the decision tree method for predicting natural gas consumption and to assess its performance through a detailed analysis. The proposed methodology has the potential to enhance the operation of district heating systems, reduce carbon dioxide emissions, and improve energy efficiency, which are critical for the development of sustainable energy networks in the future.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Predicting Gas Consumption for Energy Savings and Cost Reduction in District Heating Systems Using Decision Tree Algorithms

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Milica Tasić,
  • Ivan Ćirić,
  • Dejan Mitrovic,
  • Ana Kitić,
  • Marko Ignjatović,
  • Milica Ćirić,
  • Stevica Cvetković

摘要

This paper shows the application of supervised machine learning methods for predicting the gas consumption of the district heating system at the Faculty of Mechanical Engineering in Nis. The district heating system is automatically controlled, while its energy efficiency is continuously monitored through the SCADA system. Although the SCADA system provides detailed operational data, optimization decisions aimed at reducing energy consumption and costs are made by the heating plant operator. The focus of this research is on predicting natural gas consumption using the decision tree method, a machine learning technique, applied to a dataset encompassing the heating season from October 15, 2022, to December 31, 2022. The dataset includes various energy indicators obtained from the SCADA system of the heating plant, such as weather conditions, historical consumption, and other system parameters. MATLAB software was utilized for the development and testing of the model. The model's predictions were validated using data from October 15, 2024, to December 31, 2024 to evaluate its performance on unseen data. The primary objective of this study is to demonstrate the effectiveness of the decision tree method for predicting natural gas consumption and to assess its performance through a detailed analysis. The proposed methodology has the potential to enhance the operation of district heating systems, reduce carbon dioxide emissions, and improve energy efficiency, which are critical for the development of sustainable energy networks in the future.