Construction of a Combustion Model for a Dual-Fuel Engine Based on Hybrid Drive

To achieve real-time mapping and online optimization of the combustion process of a dual-fuel engine for long-term operation，a zero-dimensional(0-D) prediction model was proposed by combining the Wiebe function with the deep learning neural network for a biodiesel/diesel dual-fuel engine based on hybrid drive. Firstly，the parameters of the double Wiebe function were calculated using the pelican optimization algorithm(POA). Then，a parameter identification model of operating parameters and Wiebe parameters was established using convolutional neural network combined with bidirectional long short-term memory neural network(CNN-Bi-LSTMNN). The Wiebe function was combined with the deep learning neural network to simplify and reconstruct the combustion process，and finally a 0-D prediction model based on hybrid drive was established，which can further obtain the cylinder pressure curve and various performance results. The results show that the 0-D prediction model of the dual-fuel engine based on double Wiebe function combined with CNN-Bi-LSTM has good prediction accuracy and generalization. The development of the model provides a reliable digital model support for online evaluation and optimization of dual-fuel engine performances. © 2024 Chinese Society for Internal Combustion Engines. All rights reserved.