CYBERSECURITY FOR NUCLEAR POWER PLANTS - Working with Simulator's Data and Machine Learning Algorithms to Find Abnormalities at Nuclear Power Plants - A Thesis

Aug 4, 2026

Upload by Anda

Cybersecurity has the utmost importance for nuclear power plants (NPPs) as demand for clean and constant energy increases. Nations must keep NPPs secure both physically and digitally while cyber-attacks grow rapidly. This thesis analyzes cyber risk for NPPs, considers digital and future attack scenarios, and reviews law and policy requirements for cyber protection. It also evaluates how data analysis and machine learning on simulator data can support anomaly monitoring when real facility data is not shared.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages129

Reading Time5 h 25 min

AccessFree to download

Abstract

Acknowledgements

Contributors and Funding Sources

Nomenclature

Table of Contents

List of Figures

List of Tables

Introduction

Background

Cybersecurity

CYBERSECURITY FOR NUCLEAR POWER PLANTS WORKING WITH SIMULATOR'S DATA AND MACHINE LEARNING ALGORITHMS TO FIND ABNORMALITIES AT NUCLEAR POWER PLANTS A Thesis by MULKIYE SUMEN Submitted to the Graduate and Professional School of Texas A&M University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Chair of Committee, Dilma Da Silva Co-Chair of Committee, Shaheen Azim Dewji Committee Members, Martin Carlisle Eman Hammad Head of Department, Scott Schaefer August 2022 Major Subject: Computer Engineering Copyright 2022 Mulkiye Sumen ABSTRACT Cybersecurity has the utmost importance for nuclear power plants (NPPs) . Demand for clean and constant energy has increased the need and use of NPPs. Countries want to have and maintain secure NPPs both physically (well-studied area) and digitally. We live in a digital world, and cyber-attacks have skyrocketed in recent years. This study explores the cyber risk for NPPs, digital attacks, potential future attacks, international aspects, and law and policy requirements of cyber protection for nuclear power plants. With the help of data analysis and machine learning algorithms, extra monitoring can be conducted on plants' data. Data monitoring applications require comprehensive data to build models and develop solutions. However, nuclear facilities do not share their data because of security concerns. Plant simulators are heavily used for training people and conducting experiments. In this thesis, we inspect plant simulators to assess their usability by people with a technical background such as cyber experts, information technology technicians, and software developers. People responsible for protecting digital systems can benefit from the help of data analytic tools and machine learning models to detect abnormalities. We study machine learning models on simulator data to examine their potential in identifying anomalies. ACKNOWLEDGEMENTS I would like to thank my committee chairs Dr. Da Silva and Dr. Dewji, and my committee members Dr. Hammad and Dr. Carlisle, for their guidance and support throughout my studies and the covid-19 pandemic. I also want to thank my virtual classmates and the department faculty and staff for their help in protecting us from the virus. Finally, I want to thank Deniz and my family for their patience, love, and support. CONTRIBUTORS AND FUNDING SOURCES Contributors This work was supervised by a thesis committee consisting of Professor Dilma Da Silva of the Department of Computer Science and Engineering and Professor Shaheen Azim Dewji of the Department of Nuclear Engineering, Professor Martin Carlisle of the Department of Computer Science and Engineering, and Professor Eman Hammad of the Department Cybersecurity and Computer Science Texas A&M University-Commerce. All other work conducted for the thesis was completed by the student independently. Funding Sources This graduate study was supported by the Republic of Turkey Ministry of National Education. NOMENCLATURE APT Advanced Persistent Threat BOC Beginning of Cycle BWR Boiling Water Reactor CISA Computer Information Sharing Act CNC Computer Numerical Control CPS Cyber-physical Systems CRF Code of Federal Regulations DCS Distributed Control System DoS Denial of Service DT Decision Tree EOC End of Cycle HIPAA Health Insurance Portability and Accountability Act HMI Human-Machine Interface HTR Power Pressurizer Heater I&C Instrumentation and Control Systems IAEA International Atomic Energy Agency IC Integrated Circuit ICS Industrial Control Systems IED Intelligent Electronic Device IoT Internet of Things ISO International Organization for Standardization IT Information Technology KNN K-Nearest Neighbors LR Logistic Regression MBK Integrated Break Flow MDB Microsoft Access Database ML Machine Learning MOC Middle of Cycle NEA Nuclear Energy Agency NEI Nuclear Energy Institute NIST National Institute of

FAQs

为什么核电站需要优先重视网络安全？

核电站既需要持续提供清洁稳定能源，也必须同时在物理与数字层面保持安全。随着近年来网络攻击显著增加，网络安全的重要性进一步提升。

这项研究关注哪些核电站网络安全相关内容？

研究围绕核电站的网络风险展开，讨论数字攻击与潜在未来攻击，并结合网络防护的国际层面以及相关法律与政策要求。

研究如何利用模拟器数据与机器学习来发现异常？

由于核设施往往出于安全担忧不共享数据，论文检视核电站模拟器的可用性，并在模拟器数据上测试机器学习模型用于识别异常，从而为数字系统防护提供支持。

Aug 4, 2026

Upload by Anda