Validation and modeling of user interactions and their algorithmic implementation in building automation

controlled test environment Copyright: © Johannes Lang, BINE Informationsdienst

Project duration:

06/2015 to 09/2018

Project funding:

BMWi, Förderkennzeichen 03ET1289A

Project consortium:

  • Karlsruhe Institute of Technology (KIT)
  • Institute of Energy Efficiency and Sustainable Building (E3D)
  • Institute for Energy Efficient Buildings and Indoor Climate, RWTH Aachen (EBC)
  • Fraunhofer Institute for Building Physics (IBP)
  • ABB AG, Corporate Research Center Germany, Building Automation


User behavior has considerable influence on the primary energy consumption of buildings and, thus, on the energy-related emissions. Further, this influence tends to increase with enhanced energy standards. It can be observed that the energy consumption of energetically ambitious buildings shows significant differences to the anticipated demand values. So far, interaction between users and buildings is only insufficiently implemented in planning tools and in the scope of building automation. There is a lack of evidence-based algorithms – for example to describe the impact of user behavior realistically based on model-predictive control and regulation.

The aim of the sub-project “Modellierung und Validierung im Feldversuch” (practical modeling and validating) is to study user behavior with respect to interactions between humans and the building envelope in passively cooled buildings, as well as man-machine interaction in the scope of regulation. Thermal and visual comfort serve as target variables. Here, controlled experiments using an indoor climate test stand serve to study actual behavior, accompanied by field tests inside a university building for the sake of metrological verification and evaluation. The data serve as a basis for a probability-based (stochastic) model to map the behavior of people in rooms – for example to determine the probability that a user will open a window as a reaction to given environment variables (ambient air temperature, CO2 concentration, etc.). The developed model is prototypically implemented and validated in a building automation environment.

The project is implemented in close cooperation with the Institute of Energy Efficient Buildings and Indoor Climate (EBC) at RWTH Aachen, the Building Science Group

(fbta) of the KIT Karlsruhe, the Fraunhofer Institute for Building Physics (IBP), as well as the industry partner ABB. The project also includes an international cooperation in the scope of EBC Annex 66 “Definition and Simulation of Occupant Behavior in Buildings” under the aegis of the International Energy Agency (IEA).