A robust-weighted hybrid nonlinear regression for reliability based topology optimization with multi-source uncertainties

Shiyuan Yang

doi:10.1016/j.cma.2025.118360

A robust-weighted hybrid nonlinear regression for reliability based topology optimization with multi-source uncertainties

2025-09-05

0

PRE

AI

DOI：10.1016/j.cma.2025.118360

Original

Original request for help

Save

Abstract

En 中文

• A novel hybrid nonlinear learning method is proposed using improved HS optimization for TO and RBTO. • The absolute weighted bi-linear loss function applied for training nonlinear function applied in inverse TO under uncertainties • The inverse TO under multi-uncertainties computed by nonlinear and weighted nonlinear models are compared with TO-based bisection. • The computational burden with accurate TO and RBTO results is captured by hybrid weighted nonlinear models

A robust-weighted hybrid nonlinear regression for reliability based topology optimization with multi-source uncertainties

Abstract

Journal

Organization