Optimizing Operations and Maintenance Through Machine Learning - Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning

Aug 3, 2026

Upload by Skyler

Safety and reliability are of the utmost importance in safety-critical systems, such as nuclear power plants. Anomalies in these systems can arise from sensor faults, environmental factors, and human error, undermining plant integrity. If left undetected, these anomalies can cause unexpected failures, costly downtime, and potentially significant safety incidents. Effective anomaly detection therefore supports reliability and improves operation and maintenance performance by enabling timely error identification, capture, and mitigation using machine learning-based anomaly detection methods.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages126

Reading Time5 h 18 min

AccessFree to download

Acknowledgements

Doctoral journey and support

Abstract

Safety-critical systems and anomaly detection

Machine learning approaches

University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange | Doctoral Dissertations | Graduate School | | --- | --- | | 12-2024
Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning
Ezgi Gursel
University of Tennessee, Knoxville, [egursel@vols.utk.edu](egursel@vols.utk.edu)
Follow this and additional works at: [https://trace.tennessee.edu/utk_graddiss](https://trace.tennessee.edu/utk_graddiss)
Part of the Industrial Engineering Commons | | Recommended Citation Gursel, Ezgi, "Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning. " PhD diss., University of Tennessee, 2024. [https://trace.tennessee.edu/utk_graddiss/1](https://trace.tennessee.edu/utk_graddiss/1)1357 This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [trace@utk.edu](trace@utk.edu). To the Graduate Council: I am submitting herewith a dissertation written by Ezgi Gursel entitled "Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, PhysicsInformed Machine Learning, and Reinforcement Learning." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Industrial Engineering. Anahita Khojandi, Major Professor We have read this dissertation and recommend its acceptance: Anahita Khojandi, Jim Ostrowski, Bing Yao, Jamie Coble Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School (Original signatures are on file with official student records.) To the Graduate Council: I am submitting herewith a dissertation written by Ezgi Gursel entitled “Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning.” I have examined the final paper copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Industrial Engineering. Anahita Khojandi, Major Professor We have read this dissertation and recommend its acceptance: | Anahita Khojandi | | --- | | Jim Ostrowski | | Bing Yao | Jamie Coble Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School To the Graduate Council: I am submitting herewith a dissertation written by Ezgi Gursel entitled “Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning.” I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Industrial Engineering. Anahita Khojandi, Major Professor We have read this dissertation and recommend its acceptance: Anahita Khojandi | Jim Ostrowski | | --- | | Bing Yao | | Jamie Coble | Accepted for the Council: Dixie L. Thompson Vice Provost and Dean of the Graduate School (Original signatures are on file with official student records.) Optimizing Operations and Maintenance Through Machine Learning: Generative Adversarial Networks, Physics-Informed Machine Learning, and Reinforcement Learning A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Ezgi Gursel Decemb

FAQs

Why is anomaly detection critical for safety-critical systems?

Anomalies in safety-critical systems can stem from sensor faults, environmental factors, and human error. If undetected, they may trigger unexpected failures, costly downtime, and potentially significant safety incidents.

What factors can cause anomalies in nuclear power plant-like systems?

Common sources include sensor faults, environmental influences, and human error. These factors can erode plant integrity and reliability.

How does the dissertation position machine learning for anomaly detection?

It highlights machine learning models—especially anomaly detection algorithms—as promising tools to capture anomalies, enabling effective error detection and mitigation to support reliable operation and maintenance.

Aug 3, 2026

Upload by Skyler