A universal framework for finding the optimal laminate design rules using generative and explainable machine learning models

Cheng Qiu

doi:10.1016/j.coco.2025.102548

A universal framework for finding the optimal laminate design rules using generative and explainable machine learning models

2025-08-05

0

PRE

AI

DOI：10.1016/j.coco.2025.102548

Original

Original request for help

Save

Abstract

En 中文

• The data-driven framework is very versatile and can be easily extended to complex multi-objective or non-mechanical design problems. • The laminate pattern of [θ/−θ/−θ/θ/−θ/θ/θ−θ] is a novel pattern for 8-layer homogeneous laminate design, which is found by our proposed framework. • The framework is applied to problems of the quasi-isotropic and homogeneous laminate design. The generated laminating rules are validated by the analysis of classical laminate theories.

Keywords:

data-driven framework

laminate design

multi-objective optimization

quasi-isotropic laminates

classical laminate theory

A universal framework for finding the optimal laminate design rules using generative and explainable machine learning models

Abstract

Journal

Organization