Development of the Machine Learning-based Safety Significant Factor Inference Model for Diagnosis in Autonomous Control System. (November 2021)

• Record Type: Journal Article
• Authors: Lee, Joomyung; Lin, Linyu; Athe, Paridhi; Dinh, Nam
• Is Part Of: Annals of nuclear energy. Volume 162(2021)
• Publication Date: 2021-11
• Subjects: Diagnosis -- Digital twin -- Recurrent Neural Network -- Safety significant factor -- Machine Learning; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2021.108443
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: A Digital Twin for Diagnosis (DT-D) is designed to infer the plant damage states. A Safety Significant Factor (SSF) is introduced to represent the reactor states. A Recurrent Neural Network (RNN) is used to develop the SSF inference model (SSFIM). As the DT-D, the SSFIM is well-generalized, accurate, effective, and robust model. The SSFIM shows successful model performance in loss of flow accident. Abstract: As a critical component to the autonomous control system, Digital Twin for Diagnosis (DT-D) is a virtual replica of physical systems for an accurate understanding of reactor states. Since the physical damage state cannot be measured directly in transient or accident conditions, safety significant factor (SSF) is introduced as a surrogate index for physical damage states to support safety-related decision making. This study develops a machine learning (ML) based SSF inference model (SSFIM) using the Recurrent Neural Network (RNN) with acceptable accuracy, generalization capability, effectiveness, and robustness against sensor errors. To demonstrate the capability of the ML-based SSFIM, case studies are implemented on a plant simulator for Experimental Breeder Reactor – II. For partial loss of flow accident scenarios, the SSFIM is able to infer the peak fuel centerline temperature with minimally one sensor. Meanwhile the SSFIM is also found to be robust against manipulated sensor drifts and/or random noises.