Paper Conference

Proceedings of Building Simulation 2017: 15th Conference of IBPSA


Real-Time Assessment of Human Thermal Comfort Using Image Recognition in Conjunction with a Detailed Numerical Human Model

Henning Metzmacher1, Daniel Wo¨lki1, Carolin Schmidt1, Je´roˆme Frisch1, Christoph van Treeck1
1Institute of Energy Efficiency and Sustainable Building, E3D, RWTH Aachen University

Abstract: This paper introduces a real-time system for human thermal comfort assessment based on thermal image recognition, sensor fusion and co-simulation. The aim of this work is to provide a framework capable of establishing the best possible representation of a person’s thermal state while filling sensor related information gaps by the use of a detailed numerical human model. The latter is used to predict the thermal state for body locations hidden from the IR-camera (e.g. obstacles, clothing) and allows to complete the missing information for local/global thermal comfort predictions. For evaluation purposes, a manikin with flat heated sensors (FHS) simulating human skin is added to an experimental setup. The measured comfort data of all three systems is evaluated to verify the entire system. The paper utilizes a comfort index proposed in DIN EN ISO 14505-2 (2007).
Pages: 691 - 700