STEMS: Spatial-Temporal Enhanced Safe Multi-Agent Coordination for Building Energy Management

2025-10-15

AI Summary

Addressing challenges of insufficient spatiotemporal dependency modeling, lack of safety constraints, and high system complexity in multi-building energy systems, this paper proposes a spatiotemporal graph learning architecture integrating Graph Convolutional Networks (GCNs) and Transformers. Crucially, it introduces Control Barrier Functions (CBFs) into a multi-agent reinforcement learning (MARL) framework for the first time, ensuring mathematically rigorous operational safety guarantees. Evaluated on real-world datasets, the method achieves a 21% reduction in energy cost, an 18% decrease in carbon emissions, a drop in safety violation rate from 35.1% to 5.6%, and a thermal discomfort index of only 0.13. Moreover, it demonstrates strong robustness under extreme weather conditions and heterogeneous building configurations. This work establishes a verifiable, scalable paradigm for coordinated energy management that simultaneously ensures low carbon emissions, occupant comfort, and provable safety.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Multiagent Systems: Multiagent Learning
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Building energy management is essential for achieving carbon reduction goals, improving occupant comfort, and reducing energy costs. Coordinated building energy management faces critical challenges in exploiting spatial-temporal dependencies while ensuring operational safety across multi-building systems. Current multi-building energy systems face three key challenges: insufficient spatial-temporal information exploitation, lack of rigorous safety guarantees, and system complexity. This paper proposes Spatial-Temporal Enhanced Safe Multi-Agent Coordination (STEMS), a novel safety-constrained multi-agent reinforcement learning framework for coordinated building energy management. STEMS integrates two core components: (1) a spatial-temporal graph representation learning framework using a GCN-Transformer fusion architecture to capture inter-building relationships and temporal patterns, and (2) a safety-constrained multi-agent RL algorithm incorporating Control Barrier Functions to provide mathematical safety guarantees. Extensive experiments on real-world building datasets demonstrate STEMS's superior performance over existing methods, showing that STEMS achieves 21% cost reduction, 18% emission reduction, and dramatically reduces safety violations from 35.1% to 5.6% while maintaining optimal comfort with only 0.13 discomfort proportion. The framework also demonstrates strong robustness during extreme weather conditions and maintains effectiveness across different building types.

Problem

• Addresses insufficient spatial-temporal information in multi-building systems
• Ensures rigorous safety guarantees for energy coordination operations
• Reduces system complexity while optimizing cost and emissions

Innovation

• GCN-Transformer fusion captures spatial-temporal dependencies
• Control Barrier Functions ensure mathematical safety guarantees
• Multi-agent reinforcement learning coordinates building energy management