ErrorCAD

In order to optimize the control-command system of its nuclear boiler, AREVA TA must estimate confidence intervals for primary parameters (core temperature, steam generator level, neutron flux, etc.).

# ErrorCAD

In complex measuring chains (control-command), companies guarantee a level of precision (uncertainty) for physical variables measured. When several sources of random errors arise during measurement-taking, the uncertainty can be difficult to determine, so an upper bound is usually used.

In the area of reliability, the rates of reliability/safety are calculated using a fault tree and the probability of failure for each component. Reliability software determines the likelihood of a breakdown but does not calculate its dispersion.

ErrorCAD is a tool for propagating the law of error in a measuring chain or fault tree. ErrorCAD is the best alternative to a Monte Carlo simulation when there are multiple sources of risk or when the user is interested in the asymptomatic part of the law of error. ErrorCAD works by modelling the measuring chain as a graph and identifies sizing components in the overall error.

**Product sheet**