MEtmodélisation de Liaison Optimale DYnamique Amortissante

The assessment of the remaining useful life (RUL) is a crucial piece of data when considering extending the operation of infrastructure. Indeed, the preventive renewal of certain facilities imposes, in addition to the financial cost, an unjustifiable pressure on natural resources (use of raw materials and production of unnecessary waste).

Current predictive longevity models must integrate, alongside useful failure indicators for long-term durability testing, realistic failure models to reduce the need for testing.

Microstructural physical damage models, however, are difficult to implement at the scale of some structures. The size and accessibility of physical measurements are sometimes prohibitive: transportation equipment and infrastructure (suspensions, vibration isolation), energy production and transmission equipment (power line pylons).

Both modeling and the interpretation of experimental results require consideration of dynamics involving large displacements, random loads, and significant inertial effects. The study is therefore highly nonlinear given the wide variation in parameters and modeling. This requires an analysis that can incorporate a high level of uncertainty.

It is therefore necessary to generate and define simpler models that offer reasonable compromises between their representativeness and the effectiveness of their deployment in algorithms and numerical schemes. Meta-modeling plays a major role in this context.

This research aims to expand the scope of application and deployment of metamodels for structural health monitoring. The innovation lies in the definition of indicators and precursor factors of ruin based on dynamic measurements, an important topic at the national level for the maintenance of facilities.

The LMN structures its activities on the evaluation and prediction of the behavior of structures and their interactions. It combines theoretical analysis and experimental development for reliability assessment fatigue testing and structural health monitoring.