【摘要】 通过比较发动机多工况排放的台架测试与基于燃烧产物成分的化学平衡方程推导结果,表明各种以燃烧产物组分分析空燃比(A/F)的方法所得值不同。提出以神经网络非线性映射能力,建立发动机各工况点燃烧产物组分和A/F之间的相关关系,以已知信息作为神经网络的输入和输出。构建训练样本。经过对样本的学习和训练,应用其内插和外推的泛化能力。实现以燃烧产物组分来精确映射空燃比A/F。消除了各种方法映射A/F的适用条件限制,提高了整个工况范围的A/F映射检测精度,为进一步改善发动机的排放污染提供了参考依据。更多还原

【Abstract】 Three chemical equation and Bench test results were compared.Air-Fuel Ratio(A/F) deduced with combustion products components by four ways was different.Using Neural Networks(NN)mapping ability,the relation was found in combustion products components and A/F.Sample was fabricated and leamed using those known information as input and output of NN,A/F could be mapping with combustion products components for engine by generalization of interpolation and extrapolation;Prime information feature was abstracted based compared and verified with multi means.A/F can be well mapping in all work conditions with combustion products components.Each means restrict was solved and deduced precision was heightened.Hence it provided a reference foundation for improving emission of engines.更多还原