基于人工神经网络的船舶油耗模型

针对船舶能效管理计划SEEMP的实施需求，寻求建立油耗模型的通用方法，为航行优化提供决策基础。以丹麦籍客滚轮MS Smyril号作为研究案例，对船舶实测运行数据进行了分析和预处理，并采用人工神经网络构建了船舶油耗和航速的黑箱模型。基于实测数据与模型预测数据的对比，验证了上述油耗模型的准确性和实用性，对于提高船舶能效运行指数EEOI具有重要意义。

Vessel Fuel Consumption Model based on Neural Network

For the implementation of Ship Energy Efficiency Management Plan (SEEMP), a common approach is presented to establish a prediction model for vessel fuel consumption, which provides a decision basis for voyage planning optimization. Danish ferry MS Smyril has been chosen as a study case. Based on the its measured voyage data, a fuel consumption model was constructed with artificial neural network. From comparison of measured data with model predictions, the accuracy of the model was validated. It will play a significant role in improving Energy Efficiency Operational Indicator (EEOI) of all ships.