Principal Investigators 

Prof. Dr.-Ing. Wolfgang Utschick
Professorship for Methods in Signal Processing, Department of Electrical and Computer Engineering, TUM School of Computation, Information and Technology

Prof. Dr.-Ing. Prof. h. c. Christoph Gehlen
Chair for Materials Science and Testing, TUM School of Engineering and Design

Motivation  and Goals 

Early damage detection and preventive maintenance is essential to reduce costs and safeguard infrastructure. The recent partial collapse of the Carola Bridge illustrates the urgency of developing novel methods for detecting early signs of precursor events that would lead to loss of load carrying capacity of  concrete infrastructure.

The main objective of this joint project is the development of a robust methodology for identifying the state of damage in concrete (early stage damage) using Coda Signals and Machine Learning. The project is focused on the extraction and categorization of new and possibly undiscovered or underestimated features.

Recent Results 

We are still at a very early stage in the project as the the official start of the sub-project (CG) was in 15.07.2024 and for the sub-project (WU) is in 1. January 2025.  A summary of the first activities and results are listed below:

• Datasets for  coda signals for tensile and compressive loadings have been collected and preprocessed. Fig. 1 shows the coefficient of correlation (CC) for various level of stretching of the signals and load steps, showing a continuously smooth variation of the CC. It was found that the change in shape of the CC vs Stretch with loading is uniquely different for tensile and compressive loading.
• In a preliminary analysis, we have extracted features from the full wave using  Principal Component Analysis (PCA) and k-means clustering. The three colors in Fig 2. correspond to the three classes. If the clusters are manually labeled, then the PC1 scores and the corresponding labels can be used to train a classifier.
• We are planning to submit a publication by the end of Q2 2025.