AN ALTERNATIVE TO MONTE CARLO SIMULATION METHOD

JORGE BALLABEN; HÉCTOR E. GOICOECHEA; MARTA B. ROSALES

Tucumán

Congreso; MECOM 2018 : XII Congreso Argentino de Mecánica Computacional; 2018

AMCA

The quantification and propagation of uncertainty is a growing discipline, with applicationswithin practically all sciences. Uncertainties are present in every prediction model of each discipline(natural, structural, biological, etc), since an exact and perfect definition of geometry, boundary conditions, material properties, initial conditions and excitations (among others) is rarely possible. A commonand robust approach to perform the propagation of uncertainties is the Monte Carlo method, which usually implies running a large number of simulations. Complex systems, where uncertainty propagation isparticularly interesting, require time expensive computations, and large memory and storage capacitiesin order to process such amount of data. Even thousands of runs of a slightly non-linear model with afew degrees of freedom could take a considerable time, despite the use of state-of-the-art solvers andparallelization techniques. In this work, a methodology that could allow the reduction of the number ofsimulations is discussed. The idea of the method is to perform a parametric sweep for a certain parameterX to be considered stochastic, then assign probabilities (according to a previously selected cumulativeprobability density function) to the values of X, and finally map the corresponding probability values tothe target variables. Hence, the probability density function of the target variables could be estimated.Within this work, the theory and implementation of the proposed method are discussed and applicationexamples are provided.