【摘要】 建立了一个基于BP神经网络的碳烟排放模型,用它来预测电控LPG-柴油双燃料的碳烟排放。模型将转速、油泵齿条位置、油门开度、LPG喷气时间、冷却水温度、转矩作为输入变量。本文采用准牛顿反向传播算法,对试验样本数据进行训练,并对此模型进行了泛化检验,将预测结果与试验结果进行了比较。此排放模型可以指导双燃料内燃机的排放性能优化试验,减少试验工作量。更多还原

【Abstract】 Based on the BP neural network,this paper established a particulate emission model for the electronically controlled LPGdiesel dual fuel engine.This model took the parameters of engine speed,fuel pump rack position,pedal,LPG(injection) time,cooling water temperature and torque as the input variables.Using BFGS arithmetic to train the experimental data and verify the extensive ability of the model,the comparison between the prediction and experimental data was carried out.The emission model can be used to guide the experimentation of emission performance optimization and to reduce the(experimental) workload.更多还原