Application of Regression Analysis Using Broad Learning System for Time-Series Forecast of Ship Fuel Consumption

Accurately forecasting the fuel consumption of ships is critical for improving their energy efficiency. However, environmental factors that affect ship have not been researched comprehensively, and most relevant studies continue to present efficiency accuracy issues. In view such problems, a time-series model based on novel regression analysis using broad learning system (BLS) was developed in this study. The BLS compared diverse set models analyses machine techniques, including autoregressive integrated moving average with exogenous inputs (ARIMAX), support vector (SVR), recurrent neural network (RNN), long short-term memory (LSTM), extreme machines (ELM). experiment, two types passenger roll-on roll-off (ro-ro) liquefied petroleum gas (LPG) carrierwere used as research objects verify proposed method’s generalizability, data divided among groups (RM, RB). experimental results showed best choice forecast actual navigation, mean absolute error (MAE) values 0.0140 0.0115 RM RB, respectively. For LPG carrier, it has also proven effect improved when factoring sea condition, MAE reaching 0.0108 0.0142 under ballast laden conditions, Furthermore, features advantages low computing complexity short time, making more suitable real-world applications. study can therefore effectively improve by reducing operating costs emissions.