Ship Energy Consumption Evaluation for Mitigation Measures Using Back-Propagation Neural Network

Abstract

As the main mode of transportation for international trade, shipping has a large volume of transportation and low freight rate, but there are problems of large fuel consumption and large emissions. Therefore, it is necessary to take some mitigation measures to save energy and reduce emissions. Many mitigation measures have been proposed based on various factors affecting ship energy consumption. To assess the performance of these mitigation measures, the energy savings of these measures have to be evaluated. Due to the complexity of the ship energy system, these factors are of different importance and may be related each other. In this paper, several influencing factors have been chosen to assess the effects of different mitigation measures on ship energy consumption, including ship conditions (speed, draft, trim, cargo volume) and weather conditions (wind, wave). A chemical tanker is taken as the research object to analyze the ship energy system and an artificial neural network model is applied to predict and evaluate the energy consumption for different mitigation measures. Moreover, various adjustments are made to the neural network structure, and the accuracy of different structures is compared based on their prediction results. The optimal neural network structure is further identified for ship energy consumption’s prediction and evaluation.