Online autonomous calibration of digital twins using machine learning with application to nuclear power plants

Data assimilation for online model calibration in discrete event simulation

“…A machine learning-based method was developed to support online autonomous calibration of digital twin models for nuclear power plants. 18 The work of Ward et al 19 presented a particle filter-based method for continuous calibration of a digital twin model and compared its performance with static and sequential Bayesian calibration approaches. Another work by Titscher et al 20 developed a Bayesian calibration method and applied it to online model calibration using real measurement data from a lab-based demonstrator bridge.…”

A review of the application of artificial intelligence to nuclear reactors: Where we are and what's next

“…Gong and Cheng [ 94 ] presented a digital-twin-based approach to predict the high-dimensional output quantities of interest, such as the neutron flux and power distributions, by combining ML and ROM. Song and Song [ 95 ] proposed an autonomous calibration method for NPP digital twins to compensate for the computational bias of low-precision models. An initial calibration model was established at the offline end using the error database and ML methods, and the model was dynamically and continuously updated at the online end using real-time measurement data.…”

“…However, since the online BPNN still relies on time-consuming gradient descent iterative training, in order to ensure the computation speed, the number of training steps is greatly limited. [ 89 ] Online sequential condition prediction EOS-ELM Fast learning speeding without obvious overfitting problems [ 90 ] Water level prediction DNN-GA DNN model has better performance than cascaded fuzzy neural network [ 91 ] Operating parameters prediction during LOCA DNN/LSTM Proposed methods are 100,000 times faster than the original simulation tool with satisfying accuracy [ 92 ] Prediction of neutron flux and power distributions ROM-ML Able to predict high-dimensional outputs with physics-informed digital twins framework [ 94 ] Compensation for low-precision model deviation K-means and ANN A digital twin model consisting of offline and online stages is proposed, and its calibration results are shown to have good agreement with the ground truth [ 95 ] System-level FD ANN 8 operating conditions can be accurately diagnosed and classified [ 98 ] System-level FD ANN A dynamic architecture was proposed, in which the first network is used to judge whether the system is in an abnormal state, and the second network is used for classification of abnormal conditions [ 99 ] System-level FD PCA PCA enables fast compression of multiple dimensions for transient identification [ 100 ] System-level FD RBFNN Able to recognize the three accidents, even with a noise level up to 10% …”

Intelligent soft computational models integrated for the prediction of potentially toxic elements and groundwater quality indicators: a case study

“…ANN is a powerful tool designed to mimic the neural functions of the human nervous system (Wagh et al, 2017a ). As a result, ANN has the ability to learn a dataset, and its learning ability aids in simulating complex nonlinear relationships (Agatonovic-Kustrin & Beresford, 2000 ; Saljooghi & Hezarkhani, 2015 ; Song et al, 2022 ; Uncuoglu et al, 2022 ), making it possible to produce meaning out of a dataset in a short period of time. The general structure of the ANN consists of an input layer, a hidden layer, and an output layer, each with numerous neurons (Diamantopoulou et al, 2005 ; Pandey et al, 2016 ; Rai et al, 2005 ).…”