NP-ODE: Neural process aided ordinary differential equations for uncertainty quantification of finite element analysis

Finite Element Analysis (FEA) has been widely used to generate simulations of complex nonlinear systems. Despite its strength and accuracy, FEA usually two limitations: (i) running high-fidelity often requires high computational cost consumes a large amount time; (ii) is deterministic method that insufficient for uncertainty quantification when modeling systems with various types uncertainties. In this article, physics-informed data-driven surrogate model, named Neural Process Aided Ordinary Differential Equation (NP-ODE), proposed model the capture both input output To validate advantages NP-ODE, we conduct experiments on simulation data generated from given ordinary differential equation collected real platform tribocorrosion. The results show NP-ODE outperforms benchmark methods. realizes smallest predictive error as well generating most reasonable confidence intervals best coverage testing points. Appendices, code, are available in supplementary files.