Integrating Symbolic Neural Networks with Building Physics: A Study and Proposal

TL;DR

This paper investigates the use of Kolmogorov-Arnold Networks (KAN) in building physics to enhance predictive modeling, knowledge discovery, and continuous learning, demonstrating their ability to rediscover equations and model complex heat transfer dynamics.

Contribution

It introduces the application of KAN in building physics, showcasing their potential for knowledge augmentation and proposing a decision tree for model selection in this domain.

Findings

KAN can rediscover fundamental heat transfer equations

KAN effectively approximates complex formulas

KAN captures time-dependent heat transfer dynamics

Abstract

Symbolic neural networks, such as Kolmogorov-Arnold Networks (KAN), offer a promising approach for integrating prior knowledge with data-driven methods, making them valuable for addressing inverse problems in scientific and engineering domains. This study explores the application of KAN in building physics, focusing on predictive modeling, knowledge discovery, and continuous learning. Through four case studies, we demonstrate KAN's ability to rediscover fundamental equations, approximate complex formulas, and capture time-dependent dynamics in heat transfer. While there are challenges in extrapolation and interpretability, we highlight KAN's potential to combine advanced modeling methods for knowledge augmentation, which benefits energy efficiency, system optimization, and sustainability assessments beyond the personal knowledge constraints of the modelers. Additionally, we propose a model selection decision tree to guide practitioners in appropriate applications for building physics.