Effective discrepancy and numerical experiments

Many problems require the computation of a high dimensional integral, typically with a few tens of input factors, with a low number of integrand evaluations. To avoid the curse of dimensionality, we reduce the dimension before applying the Quasi-Monte Carlo method. We will show how to reduce the dimension by computing approximate Sobol indices of the variables with a two-levels fractional factorial design. Then, we will use the Sobol indices to define the effective discrepancy, which turns out to be correlated with the QMC error and thus enables one to choose a good sequence for the integral estimation.     

Study of potential leakage on several stressed fittings for hydrogen pressures up to 700 bar

In order to improve risk analyses and influence the design of the future H₂ systems, an experimental study on “real” leaks qualification and quantification was performed. In H₂ energy applications, fittings appeared as a significant leakage potential and subsequently explosion and flame hazards. Thus, as a part of the “Horizon Hydrogène Energie” French program, four kinds of commercial fittings usually employed on H₂ systems were tested thanks to a new high pressure test bench – designed, setup and operated by INERIS – allowing experiments to be led for H₂ pressures until 700 bar. The fittings underwent defined stresses representative of H₂ systems lifetime and beyond. The associated leaks – when existing – are characterized in terms of flow rate.