The objective of this proposal is to develop an advanced modeling and simulation framework for predicting fire spread. The modeling is based on Large Eddy Simulation (LES) techniques using a newly developed embedded fame spread modeling (EFSM) approach that will result in unprecedented accuracy in the prediction of turbulent fame spread. Validation and verification of this new modeling approach and dissemination of knowledge to the public will be facilitated by a strong collaboration with U.S. government and industry laboratories specializing in fire science. Specifically, the agencies involved are the Fire Science and Technology department at Sandia National Laboratories (SNL), and Factory Mutual Global Research (FM Global). These collaborations will result in: (1) a web-portal will be created for the downloading and using the EFSM library and (2) a fame spread workshop will be hosted by the University at Buffalo during the second year of the effort. All of the tools and modeling methods developed in this research endeavor will be incorporated into the classroom through a class the P.I. currently teaches on Fire Science and Safety Engineering.
Intellectual Merit:
The intellectual merit of this research is in a unified predictive computational capability to predict turbulent fame spread over charring materials. The modeling advancements for this effort are focused on: (1) a new modeling approach for predicting turbulent wall fires based on the use of EFSM, (2) a thermal model to account for charring of construction materials, (3) a new near-wall boundary model and scaling theory for computing buoyancy driven turbulent reacting boundary layers. The implementation of these models will be incorporated into an existing high-order accurate fluid-structure computational framework for simulating coupled conjugate heat and mass transfer using both direct numerical simulation (DNS) and LES. The inclusion of the models will require numerical algorithm advances for use on massively parallel computers that will result in unprecedented levels of time and spatial resolution for the prediction of turbulent flame spread. These advances include: (1) an adaptive mesh refinement procedure for computation of conjugate heat and mass transfer processes across solid-gas interfaces and (2) a new embedded cut-cell approach for computing participating radiation heat transfer in complex geometries.
Broader Impact:
The impact of this research on society is to offer scientific insight on the growth of large scale structural fires. The longer term impact of this research is to provide a high fidelity, broad-based computational tool, to predict the performance of structures from fire and a trained workforce who are able to use these tools for high performance computing. The expectation is that these efforts will provide suggestions for new performance based design approaches for insuring fire safety of critical infrastructures. In addition, the broad-based mathematical, modeling and simulation framework developed as part of this research is not unique to flame spread modeling and could also be used to analyze a wide range of energy-conversion problems which involve fluid-solid conjugate heat and mass transfer processes for reacting flows.
The severe destructive damage from large-scale fire spread has captured the attention of our nation in recent times ranging from building fire to forest fires that rage across the southwest. Characteristic of all fires is their exponential growth, which comes from the continued participation of surrounding materials as fuel in the fire event. Minimizing fire growth requires a detailed understanding of conjugate heat and mass transfer processes that define fire spread and tools which can yield quantitative predictions of fire growth. The fundamental challenge in understanding fire growth is the feedback cycle of material decomposition, fuel release, combustion and subsequent heating of new material. The attached figure illustrates this cycle showing turbulent flame spread. The goal of the proposed research is to develop an advanced multi-scale, numerical modeling and simulation framework based on the use of Large Eddy Simulation (LES) techniques for the analysis of fire spread phenomena. The subgrid scale (SGS) modeling approach proposed is based on a newly developed embedded flame spread model (EFSM) concept. The advantage of this modeling approach is the ability to obtain quantitatively accurate estimates of heat flux to solid surfaces without excessive computational costs so that it may be used in practical fire safety analysis. A unique numerical simulation framework is developed to conduct first principal simulations of flame spread to explore the validity of EFSM. This research has been largely focused on developing tools and models for flame spread over charring carbon-epoxy laminated composite. The appeal of focusing on composites vs. other charring materials (e.g., wood products) is the material properties and local morphology is well defined. The research thus far has supported 2 M.S. and 2 Ph.D. students and several undergraduate student experiences. Three main outcomes have come from this effort. The first is on the development of algorithms for fully coupled simulations of mass and heat transfer processes associated with flame spread. The algorithms allow general coupling of dynamical gas-solid interfaces using CFD descriptions of the flow and finite element (FE) descriptions of solids. The two unique features of the coupling algorithm are: 1) total conservation properties (mass, momentum, energy) are conserved across phase interfaces and 2) treatment of complex geometries. These features are essential for flame spread tracking the char formation and expansion of burning materials. Predictions of flame spread over carbon-epoxy laminates show excellent agreement to experimental data – indicating the modeling approach is sound. The second outcome focuses on the description of material response. A porous media description of charring materials is developed that accounts for both the thermal and mechanical response. Detailed comparisons to data of heat release rates, burnout times and critical heat flux for flame spread are promising. The third main outcome of this study are fully coupled numerical simulations of the simultaneous response of decomposing composite materials and reactions leading the process of flame spread. This effort is unique because it is the first time fully coupled numerical simulations have been performed of flame spread over charring materials. A fully coupled simulation is a necessary requirement for unraveling the complex coupling of mass and heat transfer processes leading to the phenomena of flame spread. Intellectual Merit: The intellectual merit of these activities will result in a new unified predictive computational capability to understand and predict turbulent flame spread over charring materials. The modeling advancements for this effort are focused on: (1) a new modeling approach for predicting turbulent wall fires based on the use of EFSM, (2) a new thermal model to account for charring of construction materials, (3) a new near-wall boundary model and scaling theory for computing buoyancy driven turbulent reacting boundary layers. Broader Impacts: The impact of this effort to society is to offer scientific insight on the growth of large-scale structural fires. The longer-term impact of this research is to provide a high fidelity, broad-based computational tool, to predict the performance of structures from fire and a trained workforce who are able to use these tools for high performance computing. The expectation is that these efforts will provide suggestions for new performance based design approaches for insuring fire safety for our nations' critical infrastructures. Fire protection engineering relies heavily on the use of prescribed standards and regulations for the design of fire resistant structures. The extension of these rigid codes outside of their intended use is extremely challenging for fire protection engineers requiring extrapolation at best. A new performance based fire protection-engineering paradigm is desirable to provide more flexibility in design. The tools developed in this research are part of this paradigm.
