Contribution to the development of a thermal comfort model using artificial neural networks to predict thermal sensation

back to overview

Type and Duration

PhD-Thesis, September 2016 until July 2019 (finished)


Institute of Architecture and Planning

Main Research

Sustainable Planning and Construction

Field of Research

Sustainable Design


An essential role in sustainable building planning plays the so-called thermal comfort of humans. The exploration of first comfort or human models has its origins in aerospace engineering and found its way into the building sector in the 1970s. The first standardised model is based on investigations in closed, thermally uniform rooms. In recent years, more unconditioned and personalised ventilation measures have become more important again, as both the energy efficiency of buildings and the individual thermal comfort could be improved. Various physiological models have been developed in recent decades to illustrate the associated effects, but there is still a lack of reliable derivation or prediction of the thermal sensation. In this dissertation, a model will be developed on the basis of a physiological-thermal human model and be using artificial neural networks to be able to make statements about the thermal perception of building users. The meaningful use of this numerical method requires a significant amount of different experimental data. For the choice of the input features as well as network architecture causal interactions of the human receptors and neural pathways should be considered.


Thermal comfort, Thermal sensation, Artificial neural networks (ANN)


  • Baumgärtner, L. (2018). Contribution to the simulation of human-building interaction using artificial neural networks to predict thermal sensation. Paper presented at the 12i3m, St. Gallen, Switzerland.

  • Baumgärtner, L. (2017). Evaluation of a Solar Thermal Glass Façade with Adjustable Transparency in Cold and Hot Climates. Paper presented at the CISBAT 2017, Lausanne, Schweiz.