Pre-conference short courses

Pre-conference short courses will be held on Monday, 07 september 2026

The conference will be preceded by a free training day for doctoral students and young researchers, focusing

on the application of data science and machine learning in acoustic emission and acoustic emission modelling,

from source mechanisms to signal recording and interpretation.

Venue : INSA de Lyon

• Participation in these courses is reserved as a priority for students and post-doc registered for the EWGAE conference.
• Pre-registration is required (included in the student’s conference registration fee)
• Depending on availability, the day may be extended to the other participants.

Programme under construction

Lecture 01 taught by Prof. Dr. Markus Sause, Director Centre for Future Production (CFP), University of Augsburg

AE measurement technology - about sensor systems, calibration and verification

The sensitivity of AE sensing systems is outstanding in the way that the physical motion we measure is in the order of atomistic dimensions. Also, the AE sensor itself is an extraordinary piece of instrumentation, as it does not always respond to the same physical quantity (e.g. displacement, velocity, acceleration or pressure). This actually depends on the way how these sensors are built. However, this is nowadays used to guide the fabrication of customized MEMS sensors. But how do we know about what is going on inside these sensors? Numerical modeling is the key to understand the fundamental principles on how these sensors operate.

Lecture 02  taught by Dr. Aurelien Doitrand, MATEIS lab, INSA of Lyon

 An Overview of Acoustic Emission Source Modeling

This lecture will provide an overview of different approaches for the numerical modeling of acoustic emission sources. It will focus on both (i) controlled sources, such as pencil lead breaks or sensor wave emission in the context of acousto-ultrasonics, and (ii) intrinsic sources, such as crack initiation and propagation in the context of fracture mechanics, in different kinds of materials, such as metals, composites, or bonded assemblies. We will discuss different ways of representing these sources and how they can be implemented in finite element calculations to further study wave propagation up to the sensors.

Lecture 03  taught by Dr. Thomas Grenier, Creatis lab, INSA of Lyon

Machine learning and acoustic emission data

This short introduction to deep learning methods will teach you the fundamentals of designing and training your own network, from data splitting to layers and well-known networks.
After this theorical introduction, there will be a hands-on session in which you will explore convolutional neural networks on an image classification problem.
For this session, you will need to bring your own laptop with a Python environment installed (installation instructions will be provided).
You do not need a computer with a GPU or high computational capacity: a simple Linux/Mac/Windows laptop is enough, and a GPU is not needed.
Do I need to be a Python programmer? No! The provided code is functional and you simply interact with it (i.e. by changing parameter values).