Improving Energy Efficiency of Oil Platforms Through Optimal Loading of Diesel Generators Using Machine Learning and Search Algorithms

A maritime oil platform in Scotland achieved significant energy efficiency improvements by optimizing the load configuration of four main diesel generators using machine learning and search algorithms. Based on data collected over the past 18 months from this platform, researchers first conducted exploratory data analysis to identify outliers, and then built a regression model to predict daily fuel consumption under different power loads. After comparing various algorithms, it was found that multiple linear regression and artificial neural networks performed best in predicting performance. On this basis, search algorithms were used to identify a generator power combination that could minimize daily fuel consumption. After implementing this optimization strategy, the platform’s average daily diesel consumption decreased by approximately 24,000 liters, a reduction of 27%.