Application of an artificial neural network for modelling the thermal dynamics of a building's space and its heating system

M. M. Gouda, Sean Danaher, Chris Underwood

Research output: Contribution to journalArticlepeer-review

34 Citations (Scopus)

Abstract

Artificial neural networks (ANNs) have been used for modelling the thermal dynamics of a building's space, its water heating system and the influence of solar radiation. A multi-layer feed-forward neural network, using a Levenberg-Marquardt backpropagation-training algorithm, has been applied to predict the future internal temperature. Real weather data for a number of winter months, together with a validated building model (based on the building constructions data), were used to train the network in order to generate a mapping between the easily measurable inputs (outdoor temperature, solar irradiance, heating valve position and the building indoor temperature) and the desired output, i.e., the predicted indoor temperature. The objective of this work was to investigate the potential of using an ANN with singular value decomposition method (SVD) to predict the indoor temperature to shut down the heating system controller early for saving the energy consumption for heating inside the building.
Original languageEnglish
Pages (from-to)333-344
JournalMathematical and Computer Modelling of Dynamical Systems
Volume8
Issue number3
DOIs
Publication statusPublished - Sep 2002

Fingerprint

Dive into the research topics of 'Application of an artificial neural network for modelling the thermal dynamics of a building's space and its heating system'. Together they form a unique fingerprint.

Cite this