Intelligent Surveillance System for DUPIC Fuel Development Facility

Year
2001
Author(s)
Won Il Ko - Korea Atomic Energy Research Institute
Ho-Dong Kim - Korea Atomic Energy Research Institute
Myung Seung Yang - Korea Atomic Energy Research Institute
Sang Yoon Lee - Korea Atomic Energy Research Institute
Dae Yong Song - Korea Atomic Energy Research Institute
Hyun Soo Park - Korea Atomic Energy Research Institute
Abstract
Unattended continuous monitoring systems result in large amounts of data, which require much time and effort to inspect. Therefore, it is necessary to develop software that automatically pinpoints and diagnoses the anomalies from the data. In this regard, this paper presents a novel concept of a continuous surveillance system that integrates visual surveillance and NDA (Non-Destructive Assay) data by the use of a neural networks based real-time diagnostic system. The intelligent surveillance system, based on pattern recognition by neural networks, is developed and its functionality is tested through the field application to the DUPIC (Direct Use of PWR fuel In CANDU) Fuel Development Facility. The integral part of the multi-sensory system and analytical paradigm may provide an effective technological alternative for safeguarding of high-level radioactive material handling facilities.