Exploiting information of fusion component tests for failure rate estimation: Divertor Inner Vertical Target component study case

Highlights: • Reliability, Availability, Maintainability and Inspectability assessment for nuclear fusion plant requires Failure Rate data. • Failure Rates for ITER Divertor Inner Vertical Target component was estimated exploiting information conveyed by qualification tests. • A Bayesian Step-Stress Accelerated Life Testing framework was adopted to infer from the likelihood function depending on the observed data. • An Arrhenius-exponential model for stress levels was included in the considered Exponential and Weibull failure distributions parameters. • The Probability of failure for considered component was derived for a simulated pattern of heat load scenario. - Abstract: Reliability and availability performance assessment are useful tools in the design and delivery of fusion components to check whether expected system behaviour is likely to be observed. These analyses rely on failure rates input data, which must be estimated for components lacking operational experience. The proposed analysis is meant to gain insight on new components Failure Rates exploiting the information conveyed by qualification tests. In particular, a model for the failure rate (Exponential and Weibull model considered) and an Arrhenius-exponential model for failure rate dependency on the stress level are assumed. A parametric expression is thus derived for considered component failure rate. A set of high flux tests data part of ITER Divertor Inner Vertical Target components qualification programme is then statistically analysed in terms of censored data set and a survival likelihood function accordingly derived. Posterior distributions for failure model unknown parameters are finally inferred to provide a first approximation failure rate estimation. Despite the problematic assumptions that the usage of such preliminary and not-tailored-on-purpose data requires, the proposed method quantitatively supports RAMI analysts in the estimation of unknown failure rates taking advantage of currently available information.