The Machine Learning for Predicting Gas Turbine Performance in Naval Vessels

Abstract

Gas turbines are essential components in modern naval vessels, providing both propulsion and power for onboard systems. However, their performance can degrade over time due to factors like fouling, erosion, and thermal fatigue, leading to increased fuel consumption and reduced operational efficiency. This paper explores the application of machine learning (ML) techniques for predicting gas turbine performance, focusing on models such as linear regression, support vector machines (SVM), random forests, and gradient boosting machines (GBM). A comprehensive literature review was conducted to assess the strengths and weaknesses of these techniques. The machine learning models were developed, fine-tuned, and evaluated using metrics such as accuracy, root mean squared error (RMSE), and R2. The results demonstrate that ensemble methods, particularly Random Forests and GBM, outperform traditional models in predicting turbine performance, offering robust, accurate, and interpretable solutions for proactive maintenance and operational optimization in naval vessels.