Paper Conference

Proceedings of Building Simulation 2017: 15th Conference of IBPSA


Simulation-time Reduction Techniques for a Retrofit Planning Tool

Georgios I. Giannakis1, Georgios D. Kontes2, Ivan Korolija3, Dimitrios V. Rovas3
1School of Production Engineering & Management, Technical University of Crete, Greece
2Technische Hochschule Nu¨rnberg & Fraunhofer IBP, Nuremberg, Germany
3Institute for Environmental Design and Engineering, University College London, UK

Abstract: The design of retrofitted energy efficient buildings is a promising option towards achieving a cost-effective improvement of the overall building sector’s energy performance. With the aim of discovering the best design for a retrofitting project in an automatic manner, a decision making (or optimization) process is usually adopted, utilizing accurate building simulation models towards evaluating the candidate retrofitting scenarios. A major factor which affects the overall computational time of such a process is the simulation execution time. Since high complexity and prohibitive simulation execution time are predominantly due to the full-scale, detailed simulation, in this work, the following simulation-time reduction methodologies are evaluated with respect to accuracy and computational effort in a test building: Hierarchical clustering; Koopman modes; and Meta-models. The simplified model that would be the outcome of these approaches, can be utilized by any optimization approach to discover the best retrofitting option.
Pages: 2014 - 2023