IMPROVING CYBER RESILIENCE OF SHIPBOARD POWER SYSTEMS USING MACHINE LEARNING

Naval vessels are increasingly implementing their own shipboard microgrids to reduce fuel consumption and to meet growing technological requirements. This improved technology comes with benefits but also creates unique risks, including exposure to cyber intrusions. To prevent exploitation of these network vulnerabilities, it is imperative that system anomalies are immediately detected. This research aims to explain how physical intrusions into shipboard components can manifest in power data and how these manifestations can be effectively detected and classified. This research uses a modified Simulink model to simulate a shipboard microgrid and various loads to create realistic test data while adhering to the DOD Interface Standard (MIL-STD-1399). This research then uses a long short term memory (LSTM) network machine learning algorithm, modeled in Python, to create a system for detecting anomalies in a shipboard microgrid. The model generates predictive data, and by comparing the predictive data to current trends, it detects outliers that result from cyber threats, as well as system component failures. This research is critical for improving the operational readiness of the fleet due to both the application of predicting component failures, and the primary objective of protecting ships from cyber threats.

Department dc:contributor.department

Electrical and Computer Engineering (ECE)

Grantor dc:publisher

Monterey, CA; Naval Postgraduate School

Year dc:date.issued

2024

Author dc:creator

• Smith, Paul F.

Advisors dc:contributor.advisor

• Oriti, Giovanna
• Thulasiraman, Preetha

Statement dc:rights

• This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.

Handle dc:identifier.uri

https://hdl.handle.net/10945/72762

OAI identifier oai:identifier

oai:calhoun.nps.edu:10945/72762