Parameters estimation in modelling of gas-gap in RBMK type reactor using Bayesian approach

Abstract

Straipsnyje pristatytas algoritmas, leidžiantis apskaiˇciuoti nestacionari u proces u (pavyzdžiui, irengini u sen˙ejimo proceso) matematini u modeli u parametr u iverˇcius bei juos patikslinti, panaudojant apriorin e informacij a ir gautus steb˙ejimo arba matavim u duomenis. Pasi¯ulytas algoritmas yra paremtas modifikuotu Bajeso metodo taikymu. Sudarytas algoritmas buvo panaudotas atliekant Ignalinos atomin˙es elektrin˙es antro energetinio bloko RBMK-1500 tipo reaktoriaus duj u tarpelio (tarp technologinio kuro kanalo ir j i supanˇci u grafito žied u) užsidarymo tikimybine analize

This study presents developed algorithm for assessment and updating estimates of parameters in the mathematical models of non-stationary processes (for instance, system ageing model, dynamic system models and so on) with respect of prior information and new obtained observations. Proposed algorithm for updating estimates of random parameters is based on modified application of Bayesian approach (BA). Developed algorithm was applied for Ignalina NPP Unit 2 RBMK-1500 reactor's closure of the gas-gap between the pressure tubes and the graphite bore probabilistic analysis