基于遗传算法优化的城市标准循环工况构建

针对传统车辆标准循环工况构建中存在的问题,提出了一种基于遗传算法的城市标准循环工况构建方法。以某城市实测行驶工况数据为依据,基于微行程分析理论,对微行程特征参数进行主成分提取及聚类,建立能够反映城市车辆实际运行的行驶工况。利用方差分析确定最佳聚类数,解决了最佳聚类数不易确定的问题。采用遗传算法对代表工况段进行优化修正,以聚类结果中的欧氏距离最小为优化目标,减小其与聚类中心的欧式距离。误差分析表明:提出的工况合成方法所生成的某城市标准循环工况特征参数平均累计误差明显减少,工况精度和一致性得到较大提高。

The Construction of Standard Driving Cycle Based on Genetic Algorithm Optimization

Presenting the construction of standard driving cycle based on genetic algorithm optimization, focusing the problems in the traditional conditions. A city was chosen as the study target, using the micro-trop theory to establish the driving cycle based on the road tests, which can reflect the driving conditions of the city actual traffic, and using principal component analyzes characteristic parameters of micro-strop and cluster analysis. Use analysis of variance to determine the optimal number of clusters, it solves the problem that the number of clusters is difficult to certain. So the minimum of Euclidean distance is the optimization goal, using genetic algorithm to fit and decrease the Euclidean distance. Error analysis shows that: use the theory presented in this paper to construct the standard driving cycle, the average accumulated error of characteristic parameters decreases greatly, the accuracy and consistency increases greatly.