基于偏最小二乘回归的柴油机微粒排放预测

摘要

The basic principle and detail algorithm steps of partial least square regression (PLSR) were introduced and the relativity between CO, HC, SOOT and PM was analyzed. Taking the CO, HC and SOOT synthetical emission as independent variables and taking the PM synthetical emission as cause variable, the PM prediction model of PLSR was established. For newly-optimized parameters, the PM emission of each condition in the ESC was predicted with the model. The results show that the difference between calculated and actual results is only 0.8 %, which proves the feasibility of the prediction model.%介绍了偏最小二乘回归算法的基本原理和详细算法步骤,分析了CO,HC和炭烟与微粒排放的相关性.以CO,HC和炭烟综合排放为自变量,以微粒综合排放为因变量,建立了基于偏最小二乘回归法的微粒预测模型.针对新优化匹配的参数,利用该模型对ESC循环每工况微粒排放进行预测.结果表明,模型预测结果与实测结果间差异仅为0.8％,验证了预测模型的可行性.