Ergonomics of the indoor environment

  EIE Copyright: © e3D

We spent most of the day indoors, be it at home or at work. Our well-being and work performance are influenced by many environmental factors. These include air temperature, lighting, CO2 concentration, but also aspects related to the work organization such as noise or interruptions. Our research in this area is guided by the belief that energy efficiency and occupants’ comfort are compatible and desirable targets. The vast energy savings needed to meet the ambitious climate goals do not have to be realized at the expense of a healthy and comfortable indoor environment, if managed in a proficient way.

In order to model the occupant behavior, we draw on Machine Learning and Deep Learning methods. Neuronal Networks are a promising approach to train prediction models with large amounts of data and outperform traditional prediction methods such as logistic regression. Numeric models like MORPHEUS help us to represent and simulate the underlying thermophysiological processes.

The interior of vehicles also represent an indoor environment which is supposed to provide maximum comfort. Local conditioning based on user feedback (e.g., heating of the steering wheel or the legroom user) allows satisfying thermal comfort in electric cars with only moderate impact on driving range. In addition, real time monitoring of driver’s skin temperature with thermal imaging enables efficient control of these conditioning strategies.

With the help of human subject studies we try to test and validate developed models, control systems, and feedback systems in realistic environments. The Urban Energy Lab 4.0 will contribute immensely to this research goal.

Our research approach is highly interdisciplinary, investigating environmental influences and occupant behavior from physiological, psychological, and behavioral perspectives.

Selected publications of our research group:

Crossing borders and methods: Comparing individual and social influences on energy saving in the United Arab Emirates and Germany
In: Energy research & social science, 90, 102561, 2022 [DOI: 10.1016/j.erss.2022.102561]

Indoor environment data time-series reconstruction using autoencoder neural networks
In: Building and environment 191, Seiten/Artikel-Nr.:107623, 2021 [DOI: 10.1016/j.buildenv.2021.107623]

Interdisciplinary parametric modelling and modularization to improve air quality, acoustics and lighting in school buildings
In: [Building Simulation Conference 2021: 17th Conference of IBPSA, BS21, 2021-08-31 - 2021-09-03, Brügge, Belgium], 2021

Personal Climatization Systems : A Review on Existing and Upcoming Concepts
In: Applied Sciences, 9 (1), 35, 2019 [DOI: 10.3390/app9010035]

Window opening model using deep learning methods
In: Building and environment, 145, 319-329, 2018 [DOI: 10.1016/j.buildenv.2018.09.024]

Learning short-term past as predictor of window opening-related human behavior in commercial buildings
In: Energy and buildings, 185, 1-11, 2018 [DOI: 10.1016/j.enbuild.2018.12.012]



+49 241 80 25033