PMMA Pyrolysis Simulation -- from Micro- to Real-Scale

Abstract: In fire spread simulations, heat transfer and pyrolysis are processes to describe the thermal degradation of solid material. In general, the necessary material parameters cannot be directly measured. They are implicitly deduced from micro- and bench-scale experiments, i.e. thermogravimetric analysis (TGA), micro-combustion (MCC) and cone calorimetry. Using a complex fire model, an inverse modelling process (IMP) is capable to find parameter sets, which are able to reproduce the experimental results. In the real-scale, however, difficulties arise predicting the fire behaviour using the deduced parameter sets. Here, we show an improved model to fit data of multiple small scale experiment types. Primarily, a gas mixture is used to model an average heat of combustion for the surrogate fuel. The pyrolysis scheme is using multiple reactions to match the mass loss (TGA), as well as the energy release (MCC). Additionally, a radiative heat flux map, based on higher resolution simulations, is used in the cone calorimeter setup. With this method, polymethylmetacrylate (PMMA) micro-scale data can be reproduced well. For the bench-scale, IMP setups are used differing in cell size and targets, which all lead to similar and good results. Yet, they show significantly different performance in the real-scale parallel panel setup.

• The paper presents a robust inverse modeling approach that integrates micro- and bench-scale experiments to improve PMMA pyrolysis simulation accuracy.
• The study employs detailed thermophysical analyses, including TGA, MCC, and a shuffled complex evolutionary algorithm, to refine material parameter estimation.
• The transition from bench to real-scale simulations highlights challenges in heat transfer and flame propagation, guiding future AI-aided model improvements.

PMMA Pyrolysis Simulation: Bridging Micro- to Real-Scale

The paper "PMMA Pyrolysis Simulation -- from Micro- to Real-Scale" by Hehnen and Arnold presents an intricate approach to simulating the pyrolysis of polymethyl methacrylate (PMMA) for fire safety engineering applications. The researchers employ a methodology that integrates experiments conducted at micro- and bench-scales and evaluates the model's performance at the real scale. This paper provides an insightful investigation into the transition of pyrolysis modeling across different scales, primarily focusing on the challenges and techniques for maintaining consistency and accuracy in fire behavior predictions.

Key Methodologies and Findings

1. Inverse Modeling Process (IMP): The authors implement an inverse modeling process using the Fire Dynamics Simulator (FDS). They establish a robust mechanism by fine-tuning various material parameters derived from thermogravimetric analysis (TGA) and micro-combustion calorimetry (MCC). The inverse modeling aims to fit small-scale experimental data accurately, adjusting parameters like thermal conductivity, specific heat and pyrolysis reactions through a shuffled complex evolutionary algorithm for global minimization.
2. Simulation Model for Micro and Bench Scales: The pyrolysis reaction scheme within the micro-scale involves a multi-reaction model of PMMA decomposition. Notable is the use of gas mixtures as surrogate fuels to more accurately replicate the heat of combustion observed in experiments, ensuring that the mass transfer from solid to gas phase aligns with realistic combustion scenarios. For the bench-scale setups, parameters are further refined through simplified cone calorimeter simulations, highlighting the necessity of accounting for uneven radiative heat fluxes and sample consumption realism.
3. Bench to Real-Scale Challenges: Transitioning from bench to real-scale reveals discrepancies attributed to the inherent complexities intrinsic to larger systems, such as heat transfer and combustion dynamics. The real-scale simulations employed module setups like a parallel panel test, encompassing large eddy simulations. However, the challenges of accurately capturing flame propagation and feedback to the panels unveil discrepancies that are associated with experimental conditions and simulation model fidelity.
4. Material Parameter Estimation: The study emphasizes the integration of surrogate fuels and gas mixture models, renouncing the common practice of single-species surrogates. This approach avoids mass scaling and enables a more direct simulation of pyrolysis products, maintaining consistency in energy translation across scales.

Implications and Future Directions

The implications of this research extend to both the theoretical understanding and practical applications of pyrolysis modeling in fire safety. By employing a comprehensive multi-scale approach, the study enhances the reliability of simulations used for safety assessments and fire prevention strategies. However, it also underscores the need for detailed parameter sensitivity analysis and model optimization to ensure accurate predictions in real-world applications. Additionally, the paper suggests incorporating medium-scale studies with ad-hoc experimental setups to better cater to specific material properties.

As a future endeavor, the authors identify a need to integrate artificial intelligence-based methods to expedite the computational intensity associated with parameter optimization. Moreover, the insights drawn from this research could lead to enhanced simulation capabilities for complex combustible materials, potentially influencing building safety design and emergency response strategies significantly.

In conclusion, this paper provides a comprehensive approach to modeling pyrolysis across different scales, addressing material parameter estimations and suggesting refinements to improve fire simulation accuracy. The challenges and methodologies presented offer a valuable basis for ongoing research and development in fire dynamics simulations.