Physics-Informed Large Language Models for HVAC Anomaly Detection with Autonomous Rule Generation

TL;DR

This paper introduces PILLM, a physics-informed large language model framework that automatically generates and refines HVAC anomaly detection rules, combining interpretability, physical plausibility, and high performance for smarter building management.

Contribution

PILLM is the first framework to embed physical principles into LLM-based anomaly detection, enabling adaptive, interpretable, and physically grounded rules for HVAC systems.

Findings

Achieves state-of-the-art anomaly detection performance.

Produces interpretable and actionable diagnostic rules.

Demonstrates effectiveness on public building datasets.

Abstract

Heating, Ventilation, and Air-Conditioning (HVAC) systems account for a substantial share of global building energy use, making reliable anomaly detection essential for improving efficiency and reducing emissions. Classical rule-based approaches offer explainability but lack adaptability, while deep learning methods provide predictive power at the cost of transparency, efficiency, and physical plausibility. Recent attempts to use Large Language Models (LLMs) for anomaly detection improve interpretability but largely ignore the physical principles that govern HVAC operations. We present PILLM, a Physics-Informed LLM framework that operates within an evolutionary loop to automatically generate, evaluate, and refine anomaly detection rules. Our approach introduces physics-informed reflection and crossover operators that embed thermodynamic and control-theoretic constraints, enabling rules that are both adaptive and physically grounded. Experiments on the public Building Fault Detection dataset show that PILLM achieves state-of-the-art performance while producing diagnostic rules that are interpretable and actionable, advancing trustworthy and deployable AI for smart building systems.