A prediction method for exhaust gas temperature of marine diesel engine based on LSTM
Yufeng Zhang, Pengpeng Liu, Xiao He, Yunpeng Jiang
Abstract
In order to improve the prediction accuracy of the exhaust gas temperature of marine diesel engine, a method for predicting the exhaust gas temperature based on LSTM network is proposed. The factors affecting cylinder exhaust gas temperature are considered simultaneously, such as engine speed, fuel system, lubricating oil system, supercharger, cylinder cooling system and air inlet system. a multi-input multi-output prediction model of the diesel engine exhaust gas temperature is established by the LSTM network, and the actual ship operating data are used to train the model and test the effectiveness of the model. The results show that LSTM is suitable for the main engine exhaust temperature prediction, and the model can reach a good prediction accuracy.