Development and assessment of prognosis digital twin in a NAMAC system. (15th December 2022)

• Record Type: Journal Article
• Authors: Lin, Linyu; Gurgen, Anil; Dinh, Nam
• Is Part Of: Annals of nuclear energy. Volume 179(2022)
• Publication Date: 2022-12-15
• Subjects: ANN artificial neural network -- DAP development and assessment process -- DT digital twin -- DT-P prognosis digital twin -- EBR-II Experimental Breeder Reactor II -- EI expected improvement -- FNN feedforward neural network -- GMM Gaussian mixture model -- LOFA loss of flow accident -- ML machine learning -- NAMAC Nearly Autonomous Management and Control -- PCC Pearson correlation coefficient -- PFCL peak fuel centerline -- PGML physics-guided machine learning -- QoI quantity of interest -- RMSE root mean squared error -- RNN recurrent neural network -- SMBO sequential model-based optimization -- TPE Tree-structured Parzen Estimator -- UQ uncertainty quantification; Digital twin -- Prognosis -- Machine learning -- Autonomous control; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2022.109439
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: Development of prognosis digital twin with LSTM recurrent network. Improvement of prognosis digital twin for better prediction and decision-making. Investigation of manual search, Bayesian approach, and physics-guided ML. Abstract: The nearly autonomous management and control (NAMAC) system is a comprehensive control system to assist plant operations by furnishing control recommendations to operators. Prognosis digital twin (DT-P) is a critical component in NAMAC for predicting action effects and supporting NAMAC decision-making during normal and accident scenarios. To quantifying and reducing uncertainty of machine-learning-based DT-Ps in multi-step predictions, this work investigates and derives insights from the application of three techniques for optimizing the performance of DT-P by long short-term memory recurrent neural networks, including manual search, sequential model-based optimization, and physics-guided machine learning. Sequential model-based optimization and physics-guide machine learning result in smallest errors when the predicting transients are similar to the training data.