Estimation of radionuclide release activity using an unscented Kalman filter

Estimation of radionuclide release is an important problem due to its impact on population and environment. Especially, radioactivity release, plume height, and wind velocity need to be estimated reliably to plan emergency response in case of any unforeseen situation. In this paper, a non-linear estimation technique based on Unscented Kalman Filter has been proposed to estimate radioactivity release, wind velocity, and height of release using environmental data collected from radiation monitors placed in the proximity of release point. The Gaussian plume model has been considered to model atmospheric dispersion phenomenon of radionuclide release and for the calculation of dose rates. The performance of the proposed estimation technique has been evaluated in terms of root mean squared error. The estimation algorithm is found to be performing satisfactorily.