Energy Efficiency Analysis Of Marine Main Engine Based On RBF Neural Network Optimized By Particle Swarm Algorithm

Due to the development of intelligent ship technology,the extensive attention has been focused on the research of ship energy efficiency status assessment.It’s beneficial to control and reduce the emission of carbon dioxide from the ship by timely detecting and judging the ship energy efficiency and it also plays a critical effect on the economic benefits of ship operation.In this study,the main engine energy efficiency of an ocean-going bulk carrier is investigated based on the radial basis function(RBF)neural network whose data comes from the real navigation conditions during sailing.Furthermore,optimization algorithm with particle swarm is applied to improve the accuracy of the RBF neural network.The research results have important significance and engineering application value for improving the ship energy efficiency evaluation system.In this paper,research work mainly consists of two parts: pattern recognition algorithm of the ship sailing status and energy efficiency evaluation algorithm of the marine diesel engine.Firstly,by analyzing the ship energy efficiency evaluation index,11 parameters reflecting the marine diesel engine and navigation environment are determined,which are used as model input after data preprocessing,and four kinds of operation mode including ship berthing mode,marine motorized navigation mode,marine constant speed navigation mode and rough sea navigation mode,are taken as output.After that,the ship sailing condition recognition model based on radial basis function neural network has been developed.The recognition rate of all samples collected in the model is 95.81%,which meets the needs of pattern recognition on board ship.In view of the relevance of navigation data,kernel principal component analysis is used to reduce the dimension of the selected 10 characteristic parameters,and 6principal components with high contribution rate are selected and retained.The energy efficiency model of marine diesel engine based on radial basis function neural network is established,and the field width of the model σ is optimized by particle swarm optimization to improve the accuracy of the model.The fuel consumption calculated in the model is compared with the actual fuel consumption,so as to evaluate the energy efficiency of marine diesel engine.The results show that the accuracy of the RBF neural network algorithm optimized by particle swarm optimization is better than that of the traditional RBF neural network algorithm,and the accuracy reaches 94.3%.Therefore,the energy efficiency model developed for marine diesel engine can meet the needs of energy efficiency evaluation on board.