Nuclear energy system’s behavior and decision making using machine learning

• DOI: 10.1016/j.nucengdes.2017.08.020
• Publication Date: 2017-09-05T20:04:31Z
• Authors: Mario Gomez Ferna...; Akira Tokuhiro; Kent Welter; Qiao Wu
• Keywords: Identification

Early versions of artificial neural networks' ability to learn from data based on multivariable statistics and optimization demanded high computational performance as multiple training iterations are necessary find an optimal local minimum. The rapid advancements in performance, storage capacity, big management have allowed machine-learning techniques improve the areas learning speed, non-linear handling, complex features identification. Machine-learning proven successful been used autonomous machines, speech recognition, natural language processing. Though application intelligence nuclear engineering domain has limited, it accurately predicted desired outcomes some instances be a worthwhile area research. objectives this study create networks topologies use Oregon State University's Multi-Application Small Light Water Reactor integrated test facility's evaluate its capability predicting systems behavior during various core power inputs loss flow accident. This uses sensors, focusing primarily reactor pressure vessel internal components. As result, able predict system with good accuracy each scenario. Its provide technical can help decision makers take actions more rapidly, identify safety issues, or intelligent potential using pattern recognition for accident identification classification. Overall, development promising industry any product processes that benefit utilizing quick analysis tool.