Inverse characterization of composites using guided waves and convolutional neural networks with dual-branch feature fusion

TL;DR

This paper introduces a dual-branch convolutional neural network approach utilizing ultrasonic guided waves to simultaneously classify composite layup sequences and regress material properties, enhancing non-destructive evaluation techniques.

Contribution

It presents a novel dual-branch CNN framework that combines classification and regression for inverse characterization of composites using guided wave data.

Findings

High accuracy in layup sequence classification

Precise estimation of material properties

Effective use of guided waves for composite analysis

Abstract

In this work, ultrasonic guided waves and a dual-branch version of convolutional neural networks are used to solve two different but related inverse problems, i.e., finding layup sequence type and identifying material properties. In the forward problem, polar group velocity representations are obtained for two fundamental Lamb wave modes using the stiffness matrix method. For the inverse problems, a supervised classification-based network is implemented to classify the polar representations into different layup sequence types (inverse problem - 1) and a regression-based network is utilized to identify the material properties (inverse problem - 2)