Skip to content
Politecnico di Milano, SIGMA Lab
All people
Portrait of Rafael Junges

Rafael Junges

Postdoc

rafael.junges@polimi.it
ORCID Scopus Google Scholar LinkedIn

Research

Rafael’s research focuses on Structural Health Monitoring (SHM) of composite structures using ultrasonic guided waves (UGWs). His work combines signal processing, physics-based methods, and machine-learning algorithms to detect, localise, and quantify structural damage from multi-sensor measurements. Particular attention is given to carbon-fibre-reinforced polymer (CFRP) structures, whose anisotropy and complex wave-propagation behaviour present challenges beyond those encountered in conventional metallic structures. These materials are widely used in sectors such as aerospace because of their high specific strength and stiffness.

PhD thesis

Towards Autonomous Structural Health Monitoring: From Supervised Learning to Generative Artificial Intelligence

Politecnico di Milano · 2026

Selected publications

  1. Unsupervised data-driven method for damage localization using guided waves

    L. Lomazzi, R. Junges, M. Giglio, F. Cadini

    Mechanical Systems and Signal Processing 208, 111038 · 2024

  2. Enhancing Lamb wave-based damage diagnosis in composite materials using a pseudo-damage boosted convolutional neural network approach

    A. Gonzalez-Jimenez, L. Lomazzi, R. Junges, M. Giglio, A. Manes, F. Cadini

    Structural Health Monitoring 23(3), 1514-1529 · 2024

  3. Mitigating the Impact of Temperature Variations on Ultrasonic Guided Wave-Based Structural Health Monitoring through Variational Autoencoders

    R. Junges, L. Lomazzi, L. Miele, M. Giglio, F. Cadini

    Sensors 24(5), 1494 · 2024

  4. Convolutional autoencoders and CGANs for unsupervised structural damage localization

    R. Junges, Z. Rastin, L. Lomazzi, M. Giglio, F. Cadini

    Mechanical Systems and Signal Processing 220, 111645 · 2024

Research in images

Damage probability maps on two composite panel geometries, from ultrasonic guided waves measured by the sensor network (circles); the white cross marks the actual damage.
Damage probability maps on two composite panel geometries, from ultrasonic guided waves measured by the sensor network (circles); the white cross marks the actual damage.
Unsupervised damage localisation: guided-wave signals are acquired, a convolutional autoencoder turns them into damage indices, and these build a damage probability map.
Unsupervised damage localisation: guided-wave signals are acquired, a convolutional autoencoder turns them into damage indices, and these build a damage probability map.