@inproceedings{29fa0ec11bfe4b5ab1829b0918290054,
title = "Multi-Stage Neural Network Framework for Probabilistic-to-Continuous Predictions",
abstract = "Machine learning tools are becoming more popular for engineering problem solving, including those in nuclear systems. Nuclear reactor systems are complex, and cannot necessarily be analytically solved or simulated without a significant amount of resources. However, high reliability models with low computational resources are desired. A new type of model, Multi-Stage Deep Neural Network (MSDNN), is proposed to create a high-reliability machine learning model by using probabilistic classification information to generate more accurate continuous predictions. Multiple MSDNN architectures are designed and compared for predicting the final steady state power of a research nuclear reactor after a power changes. Results show that MSDNN models tend to perform better than standard feedforward neural networks in terms of both accuracy and generalization. This type of model is expected to be utilized and extrapolated for future nuclear problem solving using machine learning techniques.",
keywords = "Machine Learning, Nuclear Reactor, Physics Informed, Predictive Modeling",
author = "James Daniell and Kazuma Kobayashi and Syed Alam and Ayodeji Alajo and Souvik Chakraborty and Dinesh Kumar",
note = "Publisher Copyright: {\textcopyright} 2023 American Nuclear Society, Incorporated.; 13th Nuclear Plant Instrumentation, Control and Human-Machine Interface Technologies, NPIC and HMIT 2023 ; Conference date: 15-07-2023 Through 20-07-2023",
year = "2023",
doi = "10.13182/NPICHMIT23-41198",
language = "English (US)",
series = "Proceedings of 13th Nuclear Plant Instrumentation, Control and Human-Machine Interface Technologies, NPIC and HMIT 2023",
publisher = "American Nuclear Society",
pages = "1278--1287",
booktitle = "Proceedings of 13th Nuclear Plant Instrumentation, Control and Human-Machine Interface Technologies, NPIC and HMIT 2023",
address = "United States",
}