Signal analysis and risk analysis (LASAR³)

The activities concern the development and application of methods of signal, reliability and risk analysis for fault diagnostics, prognostics, maintenance, reliability, safety, vulnerability of nuclear and other industrial components and systems.
The adopted approaches are computational and innovative, based on data (big data) and physical knowledge.

Monte Carlo simulation (subset and line sampling, particle filtering), classification, regression and prediction with artificial intelligence techniques (kriging, neural networks, deep learning, extreme learning machines, reservoir computing, support vector machine, fuzzy and spectral clustering), decision-making (portfolio decision analysis, reinforcement learning), optimization (genetic algorithms, differential evolution).