Prediction of optimum heating timing based on artificial neural network by utilizing BEMS data

Jihoon Jang, Jumi Baek, Seung Bok Leigh

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

As the energy consumption in buildings accounts for about 30% of the total energy consumption, there is a growing need for people to pay attention to energy saving in buildings. From the perspective of building life cycle, buildings consume 70–90% of energy for their operation, and therefore it is very important to reduce the energy consumed for building operation. In this regard, we developed an Artificial Neural Network (ANN) model, which predicts when the heating system should run in order to reduce energy usage on winter mornings when the energy consumption is high. Especially, existing research about heating timing uses typical variables as input data of the ANN model. But in this study, accuracy of prediction is improved by adding time variable to the ANN model. Consequently, the predictive data of the ANN model were found to be significantly similar to the empirical data of BEMS, and the prediction performance of the ANN model was approximately 13.13% of CvRMSE and 0.197% of MBE, thus satisfying ASHERAE guideline 14. Therefore, the ANN model proposed in this study would help not only in reducing the energy consumed in buildings but also in providing pleasant thermal comfort by predicting the optimal heating timing.

Original languageEnglish
Pages (from-to)66-74
Number of pages9
JournalJournal of Building Engineering
Volume22
DOIs
Publication statusPublished - 2019 Mar

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Architecture
  • Building and Construction
  • Safety, Risk, Reliability and Quality
  • Mechanics of Materials

Fingerprint Dive into the research topics of 'Prediction of optimum heating timing based on artificial neural network by utilizing BEMS data'. Together they form a unique fingerprint.

  • Cite this