基于振动信号的柴油机NO_(x)排放虚拟传感研究  

Vibration based virtual sensing of nitrogen oxide emission in CI engines

作　　者：胡桂诚 李国兴 沈亮 和超亮 杨甜甜 HU Guicheng;LI Guoxing;SHEN Liang;HE Chaoliang;YANG Tiantian(College of Mechanical and Vehicle Engineering, Taiyuan University of Technology, Taiyuan 030024,China;Sinotruk Jinan Special Vehicle Co. , Ltd. , Jinan 250000,China)

机构地区：[1]太原理工大学机械与运载工程学院,太原030024 [2]中国重汽集团济南特种车有限公司,济南250000

出　　处：《重庆理工大学学报（自然科学）》2022年第6期141-148,共8页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金青年基金项目(51805335);山西省回国留学人员科研项目(HGKY2019041)。

摘　　要：对行驶过程中车辆的NO_(x)排放进行实时监测有助于缸内燃烧过程的控制和改善,实现对排放的闭环控制和瞬态排放的有效评估,从而满足日益严苛的排放要求。提出了一种基于结构振动信号的NO_(x)虚拟传感方法。基于振动信号和缸压信号的时频图谱相似性分析方法实现了缸压信号的重构并从中提取与排放相关的信息,建立了用于预测NO_(x)的主成分回归(principal component regression,PCR)模型。在单缸柴油机试验台架上对预测模型进行了验证。结果表明:所提出的PCR模型对各种工况下的NO_(x)排放都能较好预测,与红外光谱式排放测量系统相比具有更快的响应速度。Real-time monitoring of NO_(x) emissions of vehicles during driving helps to manipulate and enhance the combustion process inside the cylinder,achieve closed-loop control of emissions,and gain an adequate evaluation of transient emissions,thereby getting more and more stringent emission requirements.This paper proposes a NO_(x) virtual sensing method based on structural vibration signals.The time-frequency variations similarity analysis method based on the vibration signal and the cylinder pressure signal realizes the reconstruction of the cylinder pressure signal.It extracts emission-related information from it,thereby establishing a principal component regression model for predicting NO_(x).The prediction model is verified on the single-cylinder diesel engine test bench.The results show that the proposed PCR model better predicts NO_(x) emissions under various working conditions and has a better response speed than infrared spectroscopy emission measurement systems.

关 键 词：虚拟传感 NO_(x)排放 振动分析 实时驾驶排放 PCR分析 

分 类 号：TK427[动力工程及工程热物理—动力机械及工程]