Energy Infrastructure Risk Modeling using Quantum and Classical Machine Learning

The integrity of energy infrastructure is critical to societal stability, yet it is increasingly vulnerable to diverse natural and man-made hazards. These hazards create interdependent risks that challenge conventional risk management models. This study explores hazard modeling approaches for energy infrastructure using classical machine learning (ML) and quantum machine learning (QML) techniques. We develop Deep Neural Networks (DNN), Naïve Bayes (NB), Decision Tree (DT), and Quantum Neural Networks (QNN) deployed in a 127-qubit quantum computer to classify hazard levels. Further, we propose a Bayesian network to model vulnerability propagation within the infrastructure. Comparing QML models against classical ML benchmarks provides insights into QML’s potential benefits and limitations for complex, high-dimensional data in critical infrastructure applications. This paper presents data analysis, model development, and comparison of classical and quantum methods to advance energy infrastructure risk modeling.