D3.3 Approximate Sequential Bayesian Filtering to Estimate Rn-222 Emanation from Ra-226 Sources from Spectra
- SMSI 2021
2021-05-03 - 2021-05-06
- SMSI 2021 - Measurement Science
- D3 Measurement Foundations II
- F. Mertes, S. Röttger, A. Röttger - Physikalisch-Technische Bundesanstalt, Braunschweig (Germany)
- 256 - 257
A new approach to assess the emanation of 222Rn from 226Ra sources based on measurements of the residual 222Rn is presented. The method incorporates the dynamics into the inference procedure, rather than resorting to previously available steady-state approximations. The algorithm is based on approximate Bayesian filtering in a switched linear dynamical system to identify regimes of changing emanation behavior from a time-series of spectral data.