Multiparameter Sensitivity Analysis of Operational Energy Efficiency for Inland River Ships Based on Backpropagation Neural Network Method

Abstract

AbstractWith the introduction of energy efficiency operational indicator (EEOI) to inland river ships, a multiparameter sensitivity analysis method was proposed to analyze the parameters affecting the operational energy efficiency of inland river ships. On the basis of experimental data, a model based on a backpropagation artificial neural network (BP-ANN) for predicting the EEOI was set up. The accuracy of this predictive model was verified. On the basis of weights and threshold values of each variable parameter gained in the trained BP-ANN, a Garson algorithm was used for calculating the parameter sensitivity factors. Results showed that, besides the engine speed, the environment conditions would also play a big part in the operational energy efficiency of inland river ships. The conclusion provides a foundation for engaging the energy efficiency improvement strategies for inland river ships.