Accurate construction of high dimensional model representation with applications to uncertainty quantification

Liu, Yaning; Hussaini, M. Y.; Ökten, Giray

doi:10.1016/j.ress.2016.03.021

Cited by 18 publications

(1 citation statement)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Emulation may be used in situations when a fire spread model has high computational time and/or a lot of simulations or calls of a given function is required, but emulators are rarely used in wildland fire research even though their potential for reducing computational time of simulations appears desirable in this field. Examples include data assimilation of a fire front via polynomial chaos (Rochoux et al, 2014); sensitivity analysis through the computation of Sobol' indices related to the area and shape of the simulated burned surface with emulation by either Gaussian processes (GP) or generalized polynomial chaos (Trucchia, Egorova, Pagnini, & Rochoux, 2019); uncertainty quantification and computation of Sobol' indices regarding the rate of spread (ROS) model of Rothermel (Rothermel, 1972) using high dimensional model representation methods (Liu, Hussaini, & Ökten, 2016); interpolation in a cell-based wildland fire spread simulator to quickly compute the values of correction factors in the relationship between advection velocity and spread angle on the basis of pre-computed values obtained in a few given configurations using a Radial Basis Function (RBF) approach (Ghisu, Arca, Pellizzaro, & Duce, 2015). Another example outside the scope of fire spread is the emulation of some outputs of a fire emission model with GP (Katurji et al, 2015).…”

Section: Introductionmentioning

confidence: 99%

Emulation of wildland fire spread simulation using deep learning

2021

View full text Add to dashboard Cite

Numerical simulation of wildland fire spread is useful to predict the locations that are likely to burn and to support decision in an operational context, notably for crisis situations and long-term planning. For short-term, the computational time of traditional simulators is too high to be tractable over large zones like a country or part of a country, especially for fire danger mapping.This issue is tackled by emulating the area of the burned surface returned after simulation of a fire igniting anywhere in Corsica island and spreading freely during one hour, with a wide range of possible environmental input conditions. A deep neural network with a hybrid architecture is used to account for two types of inputs: the spatial fields describing the surrounding landscape and the remaining scalar inputs.After training on a large simulation dataset, the network shows a satisfactory approximation error on a complementary test dataset with a MAPE of 32.8%. The convolutional part is pre-computed and the emulator is defined as the remaining part of the network, saving significant computational time. On a 32-core machine, the emulator has a speed-up factor of several thousands compared to the simulator and the overall relationship between its inputs and output is consistent with the expected physical behavior of fire spread. This reduction in computational time allows the computation of one-hour burned area map for the whole island of Corsica in less than a minute, opening new application in short-term fire danger mapping.

show abstract