基于改进增量模型的汽车行驶工况构建OA

To match the actual driving characteristics of light-duty vehicles,this study proposes a method for constructing vehicle driving conditions based on multi-source heterogeneous data fusion technology,using driving trajectory data of light-duty vehicles in real road environments.Firstly,a multi-stage data processing framework is constructed to generate a high-precision training dataset.Secondly,based on the clearly defined kinematic fragments and their velocity constraints,the appropriate kinematic segments are extracted and 16 key kinematic feature parameters are identified.Through the coupled principal component analysis-cluster analysis strategy,the principal components are extracted by the Kaiser criterion,and the improved incremental model is proposed to finally construct the driving condition curves with significant representativeness.The research results show that the cumulative contribution variance value of the proposed improved incremental model was reduced to 0.122,which fully demonstrates the accuracy and rationality of the constructed driving condition model.