A Data-Driven Intelligent Energy Efficiency Management System for Ships

Abstract

Shipping carries most of the cargo in international trade, with the fuel cost of ships is the major expense. Much attention has been paid to saving fuel in ship operation. First, to save fuel, an intelligent energy efficiency management system is developed to collect data related to ship energy efficiency. The main engine is the primary fuel-consuming equipment on a ship. It is important to carry out a comprehensive evaluation on the energy efficiency of the main engine. To achieve this goal, an ensemble machine learning approach combining the physics-based empirical model is used to perform a regression analysis on two different indicators. The indicators include the engine shaft power and fuel consumption rate per unit time. Second, the influence of the input features on energy efficiency indicators is analyzed, and some conclusions are drawn. On this basis, a novel trim optimization method based on piecewise regression is suggested, which is helpful to reduce ship resistance. In conclusion, this system can help ship operators to judge whether the main engine and ship are in good working condition and take appropriate measures to save energy for ships.