Model Form Calibration in Drift Diffusion Simulation Using Fractional Derivatives

Abstract

In modeling and simulation, modelform uncertainty arises from the lack of knowledge and simplification during the modeling process and numerical treatment for ease of computation. Traditional uncertainty quantification (UQ) approaches are based on assumptions of stochasticity in real, reciprocal, or functional spaces to make them computationally tractable. This makes the prediction of important quantities of interest, such as rare events, difficult. In this paper, a new approach to capture modelform uncertainty is proposed. It is based on fractional calculus, and its flexibility allows us to model a family of nonGaussian processes, which provides a more generic description of the physical world. A generalized fractional Fokker–Planck equation (fFPE) is used to describe the driftdiffusion processes under longrange correlations and memory effects. A new modelcalibration approach based on the maximum mutual information is proposed to reduce modelform uncertainty, where an optimization procedure is taken.