Leveraging AI modelling for FDS with Simvue: monitor and optimise for more sustainable simulations

2025-09-30

AI Summary

High computational cost and energy consumption of fire dynamics simulations hinder large-scale, high-frequency risk analysis. Method: This paper proposes an AI-driven, efficient, and sustainable simulation framework that replaces traditional CFD solvers with a customized machine learning surrogate model for accelerated prediction of thermal propagation dynamics—achieving speedups of several orders of magnitude. The framework integrates guided optimization strategies and a lightweight predictive model within the Simvue platform, incorporating simulation orchestration, data lineage tracking, and redundancy suppression mechanisms. Contribution/Results: In hazardous fire source localization tasks within buildings, the framework reduces simulation demand by a factor of ten, significantly improving energy efficiency and data reuse. It establishes a scalable, green high-performance paradigm for fire risk assessment, enabling sustainable, computationally tractable safety analysis.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

There is high demand on fire simulations, in both scale and quantity. We present a multi-pronged approach to improving the time and energy required to meet these demands. We show the ability of a custom machine learning surrogate model to predict the dynamics of heat propagation orders of magnitude faster than state-of-the-art CFD software for this application. We also demonstrate how a guided optimisation procedure can decrease the number of simulations required to meet an objective; using lightweight models to decide which simulations to run, we see a tenfold reduction when locating the most dangerous location for a fire to occur within a building based on the impact of smoke on visibility. Finally we present a framework and product, Simvue, through which we access these tools along with a host of automatic organisational and tracking features which enables future reuse of data and more savings through better management of simulations and combating redundancy.

Problem

• Predict heat propagation faster than CFD using machine learning surrogate models
• Reduce required simulation count through guided optimization procedures
• Provide framework for sustainable simulations with organizational and tracking features

Innovation

• Machine learning surrogate model predicts heat propagation dynamics
• Guided optimization reduces required simulations tenfold
• Simvue framework enables simulation management and data reuse