Prediction Model of Inland Ship Fuel Consumption Considering Influence of Navigation Status and Environmental Factors

Abstract

The strategy of ecological priority and green development made the fuel consumption of inland ships have received unprecedented attention. Fuel consumption prediction of inland ships can provide decision support for navigation planning and energy supervision. This paper takes the ships sailing on the Yangtze River trunk line as the research object, first of all, the navigation data is collected by the multi-source sensor. And then, consider the comprehensive influence of status monitoring data and environmental factors, the improved artificial neural network (ANN) is tailored to build the fuel consumption prediction model based on real-time monitoring data and environmental data. Finally, the constructed prediction model is analyzed and verified by a large amount of measurement data, and its performance of fuel consumption prediction is proved by comparing it with the traditional regression models.