摘要
汽油机空燃比控制常建立在进气道油膜动态辨识的基础上,难以从宏观上考虑发动机运行参数对缸内空燃比的影响.利用子空间辨识N4SID算法研究发动机多参数下的空燃比动态模型.由于N4SID算法辨识出的模型的精度比较差,进一步研究利用预报误差法(PEM)优化子空间N4SID算法辨识出的模型,提高模型精度.结果表明,N4SID方法计算效率较高,但辨识精度较差.利用PEM方法优化N4SID算法所计算出的结果,提高了模型精度,并保持了较高的辨识效率.
Modeling of air/fuel（A/F） ratio control of gasoline engine is usually based on fuel film dy- namic models of inlet identification, but it is hard to cover the impact of engine operation parameters on A/F ratio from the macro aspect. The N4SID algorithm is used to identify the A/F dynamic model under MISO conditions. Meanwhile, the N4SID algorithm through the PEM method is used to optimize the model. Results show that, the N4SID method is efficient while the identification accuracy is poor. PEM method is used to optimize the results from the N4SID algorithm, and improves the accuracy of the model while maintaining a high identification efficiency.
出处
《内燃机学报》
EI
CAS
CSCD
北大核心
2012年第3期248-253,共6页
Transactions of Csice
基金
内燃机燃烧学国家重点实验室开放课题资助项目(K2010-12)
广东省部产学研结合资助项目(2010A090200010)
作者简介
李顶根，博士，副教授，E-mail：lidinggcn@sina．com．
