analiza aiqbench
The paper compares two neural-network surrogates for predicting indoor airflow and temperature fields, one encoding room size through a fixed "canvas" and one using normalized coordinate channels, with both built on U-Net autoencoders. Its evidence is limited to simplified rooms with varying heights and lengths, so it shows relative accuracy against CFD on a controlled benchmark, not readiness for real building projects. For ML readers, the useful takeaway is that how geometry is encoded for a CNN surrogate affects generalization across input sizes.
Comparative Study of Canvas- and Coordinate-Based Geometry-Conditioned Neural Networks for Real-Time Prediction of Indoor Environments with Varying Room Dimensions
arXiv:2610.15885v1 Announce Type: new Abstract: Computational fluid dynamics (CFD) is computationally expensive for iterative building design. Deep learning surrogates offer faster predictions, but generalization across varying room dimensions remains challenging. This study compares two geometry-conditioned neural network surrogate models for rapid indoor airflow and temperature field prediction under varying room sizes. During the architectural design stage, this method is relevant for real-time assessment of whether conditioned supply air will effectively provide fresh and cool air throughout rooms with varying heights and lengths. The canvas-based neural network (CaNN) explicitly encodes room size by preserving different occupied regions within a fixed canvas, whereas the coordinate-based neural network (CoNN) uses normalized coordinate channels. Both employ U-Net autoencoders to separately predict airflow and temperature fields. Evaluation uses simplified rooms with varying heights and lengths as a controlled benchmark. The results show that CaNN achieves higher agreement with CFD results than CoNN, demonstrating its potential to support performance-driven building design.
Abstract:Computational fluid dynamics (CFD) is computationally expensive for iterative building design. Deep learning surrogates offer faster predictions, but generalization across varying room dimensions remains challenging. This study compares two geometry-conditioned neural network surrogate models for rapid indoor airflow and temperature field prediction under varying room sizes. During the architectural design stage, this method is relevant for real-time assessment of whether conditioned supply air will effectively provide fresh and cool air throughout rooms with varying heights and lengths. The canvas-based neural network (CaNN) explicitly encodes room size by preserving different occupied regions within a fixed canvas, whereas the coordinate-based neural network (CoNN) uses normalized coordinate channels. Both employ U-Net autoencoders to separately predict airflow and temperature fields. Evaluation uses simplified rooms with varying heights and lengths as a controlled benchmark. The results show that CaNN achieves higher agreement with CFD results than CoNN, demonstrating its potential to support performance-driven building design.
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2610.15885 [cs.AI]
(or arXiv:2610.15885v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.15885 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 247 cuvinte
