Inspection of elongated structures

The theme "Monitoring of elongated structures" focuses on objects such as prestressed cables, anchor tendons, pipelines, dikes, or stratified media. The main objective is to identify, locate, and, in some cases, characterize the damages or anomalies present in these structures. From a wave propagation perspective, these structures behave as waveguides for mechanical waves, allowing for long-distance propagation with minimal energy loss. This property offers the advantage of inspecting vast areas with a limited number of sensors or accessing hard-to-reach areas. However, the multimodal and dispersive nature of guided waves complicates their exploitation. To overcome these challenges and optimize inspection techniques using these waves, the team is developing experimental devices implemented both in the laboratory and in situ, while also developing analytical and numerical modeling tools specifically dedicated to the propagation of guided ultrasonic waves.

In the coming years, the team intends to continue this research by expanding its areas of expertise to new sectors, such as rail inspection, integrating new types of sensors, including electromagnetic sensors (EMAT), and developing advanced modeling techniques, for example, for cables with complex architectures. On the digital side, an open-source code recently made public for modeling waveguides using finite elements will be used. This will enable the simulation of structures with arbitrary cross-sectional geometry, potentially damaged (rails, cables), as well as the propagation of surface waves for applications in geophysics (dikes, stratified media). Inverse approaches for defect imaging will also be integrated.

Furthermore, monitoring slender structures is also addressed through electromagnetic waves, with an example being the location and characterization of pipelines. The main challenges lie in the development of hybrid approaches, combining physical methods from 3D radar signal processing (through migration and waveform inversion) with statistical techniques based on artificial intelligence (involving dedicated databases).