Skip to Main Content
Skip Nav Destination
Purpose

This study aims to model and optimize the strain concentration factor (SeCF) in orthotropic carbon/epoxy composite plates with centrally located countersunk holes under uniaxial tensile loading. The goal is to accurately predict SeCF values using artificial neural networks (ANNs) and to identify configurations that minimize strain concentration.

Design/methodology/approach

A comprehensive three-dimensional finite element analysis (FEA) is carried out to evaluate SeCF under varying geometric and material parameters, including hole radius-to-width ratio (r/w), plate thickness-to-hole radius ratio (t/r), countersink depth-to-thickness (Cs/t) ratio, countersink angle (θc) and laminate ply orientation (θp). An ANN is trained on FEA-generated data, with optimization procedure involving 12,000 network configurations considering different learning algorithms, activation functions and neuron counts. The optimal ANN architecture is then achieved and thoroughly validated with additional FEA data.

Findings

The optimal ANN configuration achieved uses a single hidden layer with twenty-nine neurons, the Bayesian Regularization algorithm and the Elliot symmetric sigmoid activation function (elliotsig), and has a root mean square error of 0.0010. The ANN-predicted SeCF closely match FEA results, with 95% of predictions within 1% errors. An optimized configuration of geometric and material parameters yielded a minimum SeCF of (Kt.ϵ) = 1.9634) for a plate with (r/w = 0.1), (t/r = 0.5), (Cs/t = 0.1), (θc= 80°) and (θp = 60°). This optimal configuration is confirmed with FEA and resulted in a 0.51% error.

Originality/value

SeCFs play a critical role in the structural integrity of engineering structures. Rivets, commonly used joining components, usually leave geometric discontinuities footprints in the form of countersunk holes. In fact, this problem has not been investigated in literature yet. Therefore, the present work uniquely integrates detailed FEA with a rigorously optimized ANN framework to model SeCFs in orthotropic plates with countersunk holes. The findings of this work provide valuable insights for designing orthotropic plates with reduced strain concentrations, enhancing their performance and structural integrity.

Licensed re-use rights only
You do not currently have access to this content.
Don't already have an account? Register

Purchased this content as a guest? Enter your email address to restore access.

Please enter valid email address.
Pay-Per-View Access
$41.00
Rental

or Create an Account

Close Modal
Close Modal