基于混合驱动的双燃料发动机燃烧模型的构建  

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

作　　者：胡登 王贺春[1] 杨传雷[1] 王银燕[1] Hu Deng;Wang Hechun;Yang Chuanlei;Wang Yinyan(College of Power and Energy Engineering,Harbin Engineering University,Harbin 150001,China)

机构地区：[1]哈尔滨工程大学动力与能源工程学院,黑龙江哈尔滨150001

出　　处：《内燃机学报》2024年第5期403-411,共9页Transactions of CSICE

基　　金：国家自然科学基金资助项目(52171298).

摘　　要：为实现对处于长时间运行的双燃料发动机燃烧过程的实时映射和在线优化,将Wiebe方程与深度学习神经网络相结合,提出了基于混合驱动的生物柴油/柴油双燃料发动机零维(0-D)预测模型.首先通过鹈鹕算法(POA)对双Wiebe方程参数进行求解;然后利用卷积神经网络结合双向长短时记忆神经网络(CNN-Bi-LSTMNN)建立运行参数和Wiebe参数的参数辨识模型.将Wiebe方程与深度学习神经网络相结合对燃烧过程进行简化、重构,建立基于混合驱动的零维预测模型,进而可以正向求出缸内压力曲线和各种性能结果.结果表明:基于双Wiebe方程结合CNN-Bi-LSTM构建的双燃料发动机零维预测模型具有良好的预测精度和泛化性.模型的开发为双燃料发动机性能的在线评估和优化提供了可靠的数字模型支持.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 dualfuel 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.

关 键 词：双燃料发动机 生物柴油 深度学习 燃烧模型 

分 类 号：TK414.31[动力工程及工程热物理—动力机械及工程]