A forthcoming Special Issue of the Journal of Building Performance Simulation will explore the interface between existing simulation methods and machine learning approaches, including surrogate modelling and related topics. It specifically focusses on building design, building on a previous Special Issue 'Data-driven approaches to building simulation for enhanced building operation and grid interaction' that focussed on operation.
Topics of interest include:
* Surrogate modelling (a machine learning model fitted to simulation outputs to approximate the performance of the detailed model)
* Applications of surrogate models or other ML-related methods, including in the areas of building stock modelling, design exploration tools and modelling of existing buildings.
* Physics-informed machine learning, where physical constraints are incorporated into machine learning algorithms directly
* Calibration or optimization of simulation models using machine learning approaches (note that pure optimization using e.g. genetic algorithms with simulations directly are not in scope)
* Any other ways of combining machine learning with more traditional physics-based simulations
* Developments in machine learning to aid in integration with simulation tools, for example in time series modelling
* The challenges of integrating machine learning with simulation, including assessing model accuracy, concerns of bias and gaining confidence from users
For more information see here
Abstracts should be submitted using this form
Important dates:
* Abstract submission deadline (300 words to Guest Editor): 30 September 2023
* Full-length article submission deadline: 31 December 2023
* Full-length articles can be submitted as soon as the abstract is accepted, and will be published as soon as the normal review process is complete.
Ralph Evins, Guest Editor
Dr Ralph Evins (He/Him)
Associate Professor
Energy in Cities group
ECS 422, Department of Civil Engineering
T +1 250-472-5845
Google Scholar
We acknowledge with respect the Lekwungen peoples on whose traditional territory the university stands and the Songhees, Esquimalt and WS?NE? peoples whose historical relationships with the land continue to this day.
