[231] Localisation and mass estimation of loose parts in nuclear power plant using FEA-based big data

Seongin Moon, Seongjin Han, To Kang and Soonwoo Han

Korea Atomic Energy Research Institute, 989-111 Daedeok-daero, Yuseong-gu, Daejeon 34057, Republic of Korea  

A loose-part monitoring system (LPMS) is used to detect loose parts in the reactor coolant system in nuclear power plant. Loose-part signal theory can be used to identify the impact source by using the amplitude and frequency content of detected acceleration signals. However, due to the geometric complexity of structures and components, the impact signals are distorted and it is difficult to characterise the impact source. Recently, model-based diagnostics has emphasised the importance of a model describing the behaviours of a mechanical system or component. Also, thanks to increasing computing power, the finite element analysis (FEA) method recently became an available option to calculate reliable impact response behaviour. In this paper, an FEA model to simulate the impact wave propagation behaviour, generated by a metal ball (simulated loose-part) impact, is validated by performing impact tests and the corresponding finite element analyses for a plate. A novel methodology based on FEA was proposed to estimate the mass and the impact location of a loose part and then the usefulness of the methodology was validated through a series of blind tests. It is expected that the proposed methodology can be utilised in model-based diagnostics for estimation of impact parameters such mass, velocity and impact location of a loose part in nuclear power plant.
Keywords: loose part, metal impact, mass estimation, localisation, FEA, big data.