Rapid Prediction and Parameter Evaluation of Process-Induced Deformation in L-Shape Structures Based on Feature Selection and Artificial Neural Networks
Open Access
- 3 November 2024
- journal article
- research article
- Published by MDPI AG in Journal of Composites Science
- Vol. 8 (11) , 455
Abstract
The process-induced deformation (PID) during the manufacturing of thermosetting composite materials can significantly compromise manufacturing precision. This paper introduces an innovative method that combines a finite element analysis (FEA), feature classification algorithms, and an Artificial Neural Network (ANN) framework to rapidly predict the PID of a typical L-shaped structure. Initially, a comprehensive range of parameters that influence PID are compiled in this research, followed by the generation of a dataset through FEA considering viscoelastic constitutive models, validated by experimental results. Influential parameters are classified using Random Forest and LASSO regression methods, with each parameter rated according to its impact on PID, delineating their varying degrees of importance. Subsequently, through a hyperparameter analysis, an ANN framework is developed to rapidly predict the PID, while also refining the assessment of the parameters’ significance. This innovative approach achieves a computational time reduction of 98% with less than a 5% loss in accuracy, and highlights that under limited computational conditions, considering only a subset or all of the parameters—the peak temperature, corner angle, coefficient of chemical shrinkage, coefficient of thermal expansion, curing pressure, and —minimizes accuracy loss. The study demonstrates that machine learning algorithms can effectively address the challenge of predicting composite material PID, providing valuable insights for practical manufacturing applications.
Keywords
Funding Information
- National Key Research and Development Program of China (2021YFB3401700, 2021QNRC001, 2022Z039051001)
- Young Elite Scientists Sponsorship Program by CAST (2021YFB3401700, 2021QNRC001, 2022Z039051001)
- Chinese Aeronautical Establishment (2021YFB3401700, 2021QNRC001, 2022Z039051001)
This publication has 41 references indexed in Scilit:
- Rapid prediction and inverse design of distortion behaviors of composite materials using artificial neural networksPolymers for Advanced Technologies, 2020
- Identification of process-induced residual stresses in 3D woven carbon/epoxy composites by combination of FEA and blind hole drillingComposites Part A: Applied Science and Manufacturing, 2020
- A Large-scale Finite Element Model on Micromechanical Damage and Failure of Carbon Fiber/Epoxy Composites Including Thermal Residual StressApplied Composite Materials, 2017
- A Comparison of Curing Process-Induced Residual Stresses and Cure Shrinkage in Micro-Scale Composite Structures with Different Constitutive LawsApplied Composite Materials, 2017
- Meta-modeling of the curing process of thermoset matrix composites by means of a FEM–ANN approachComposites Part B: Engineering, 2014
- Autoclave cure simulation of composite structures applying implicit and explicit FE techniquesInternational Journal of Mechanics and Materials in Design, 2012
- Role of Tool-Part Interaction in Consolidation of L-Shaped Laminates during Autoclave ProcessApplied Composite Materials, 2011
- Role of tool-part interaction in process-induced warpage of autoclave-manufactured composite structuresComposites Part A: Applied Science and Manufacturing, 2010
- Closed-form solution for process-induced stresses and deformation of a composite part cured on a solid tool: Part I – Flat geometriesComposites Part A: Applied Science and Manufacturing, 2008
- Prediction of shape distortions Part I. FE-implementation of a path dependent constitutive modelComposites Part A: Applied Science and Manufacturing, 2004