摘要
针对氧传感器对汽油机排放和催化转化器效率的影响,建立了径向基函数网络(RBFNN)和自适应神经网络模糊系统(ANFIS)相结合的汽油机排放模型。利用RBFNN的非线性逼近能力,对不考虑氧传感器劣化的汽油机排放性能进行了预测。根据汽油机排放受氧传感器劣化的影响,应用ANFIS系统对RBFNN的汽油机排放预测结果进行了修正,并预测了氧传感器劣化曲线。基于RBF-ANFIS融合预测策略,进行了汽油机负荷性能和催化转化器转化效率试验。结果表明:所设计的汽油机排放模型合理,验证了该融合预测策略具有较好的分辨率,可用于氧传感器在线劣化预测。
In order to study the influence of oxygen sensor deterioration on gasoline engine emission and catalytic converter effenciency, an emission model of gasoline engine was developed by combining Radial Basis Function （RBF） neural network with Adaptive Neural Fuzzy Inference System （ANFIS）. In this model, the nonlinear approaching capacity of the RBF network was used to forecast the gasoline engine emission without taking account of the oxygen sensor deterioration. With taking account of the influence of oxygen sensor deterioration on gasoline engine emission, the ANFIS system was used to modify the predicted results of gasoline engine emission obtained by using the RBF network so as to acquire the oxygen sensor deterioration curve. The tests of the catalytic converter efficiency and the gasoline engine performance were conducted on a vehicle based on the RBF-ANFIS prediction strategy. The test results show that the emission model of gasoline engine is reasonable, the RBF-ANFIS forecasting strategy has a better predictive function and can be used for the on-line forecast of oxygen sensor deterioration.
出处
《内燃机工程》
EI
CAS
CSCD
北大核心
2009年第5期78-82,共5页
Chinese Internal Combustion Engine Engineering
基金
国家自然基金项目(50376021
50776042)
河南省教育厅自然科学研究计划项目(2008A470008)
江苏省青蓝工程资助项目
