Electric Consumption Forecast for Ships Using Multivariate Bayesian Optimization-SE-CNN-LSTM

Many studies on reducing greenhouse gas emissions from ships have been conducted to reduce environmental pollution. Reducing the fuel oil consumption of traditional and green is a key focus these studies. The ship depends electric loads. Thus, power load estimation necessary develop methods for ships. However, data accessibility low, limiting number relevant This study proposes model estimating actual using squeeze excitation (SE), convolutional neural network (CNN), long short-term memory (LSTM). load, generated by generator, reefer container, rudder angle, water speed, wind angle were measured in 10-minute increments approximately 145 d. existing parallel direct CNN-LSTM models used evaluate performance proposed model. had lowest root mean square error (RMSE), absolute (MAE), percentage (MAPE), demonstrating best compared models.