摘要: This paper proposes an open-source Python toolbox for generalized polynomial chaos expansion of dynamic systems with initial conditions and parametric uncertainties represented by stochastic differential equations. The toolbox sup... 展开 This paper proposes an open-source Python toolbox for generalized polynomial chaos expansion of dynamic systems with initial conditions and parametric uncertainties represented by stochastic differential equations. The toolbox supports Gaussian and Uniform distribution of uncertainties and can carry out uncertainty quantification in problems subject to uncertainties with mixed probability distribution types. It has an in-built mixed sparse grid-based quadrature technique to generate collocation nodes and weights and uses the pseudospectral collocation technique to compute the coefficients of gPC expansion. Further, the statistics of the solution of the stochastic differential equation can be readily calculated using the proposed toolbox. In this paper, we demonstrate the efficacy of the proposed toolbox as compared to the traditional Monte Carlo technique. To that end, we consider numerical examples of a mass spring damper system and a fixed-wing unmanned aerial vehicle system. 收起