考虑能耗的进出站驾驶风格分类及识别模型

doi:10.13229/j.cnki.jdxbgxb.20211000

摘要/Abstract

摘要：

为实现对进出站驾驶风格的分类与识别，基于纯电动公交车自然驾驶过程中的进出站片段数据，选取14个驾驶行为表征指标，利用主成分分析法对指标降维，建立K-means聚类模型将进出站片段聚为3类；以经济性、动力性和舒适性为语义解释的3个维度，对3类群集语义解释为耗能激进型、一般型和节能舒适型；建立三层BP神经网络模型，实现进出站驾驶风格的在线识别。模型验证发现，识别模型各评价指标值均处于0附近，且模型平均识别率为93.52%，可以较好地实现对任意进出站片段的驾驶风格识别。

关键词: 交通工程, 驾驶行为, 聚类, 在线识别模型, 驾驶风格, 能源经济性

Abstract:

To realize the classification and recognition of entering and leaving stops′ driving styles， based on the entering and leaving stops data in the natural driving process of pure electric bus， 14 driving behavior characterization indexes were selected， the dimension of the indexes is reduced by principal component analysis， and a K-means clustering model was established to cluster the entering and leaving stops segments into three categories. Taking economy， dynamic and comfort as the three dimensions of semantic interpretation， the three types were interpreted as high energy consumption & aggressive style， general style and energy-saving & comfort style. A three-layer BP neural network model was established to realize the on-line recognition of driving style. The model verification showed that the evaluation index values of the recognition model are near 0， and the average recognition rate of the model is 93.52%， which can better realize the driving style recognition of any entering and leaving stops segment.

Key words: traffic engineering, driving behavior, clustering, online identification model, ecological driving style, energy economy

中图分类号:

U121

张雅丽,付锐,袁伟,郭应时. 考虑能耗的进出站驾驶风格分类及识别模型[J]. 吉林大学学报(工学版), 2023, 53(7): 2029-2042.

Ya-li ZHANG,Rui FU,Wei YUAN,Ying-shi GUO. Classification and recognition model of entering and leaving stops' driving style considering energy consumption[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(7): 2029-2042.

图/表 21

图1

表1

行驶片段统计值"

指标名称	符号	计算公式
速度均值	$v ˉ$	$v ˉ = 1 n ∑ i = 1 n v i, v i ≠ 0$
速度标准差	$S v$	$S v = ∑ i = 1 n (v i - v ˉ) 2 n - 1$
加速踏板开度均值	$θ ˉ 1$	$θ ˉ 1$ $= 1 n ∑ i = 1 n θ 1 i$
加速踏板开度标准差	$S θ 1$	$S θ 1 = ∑ i = 1 n (θ 1 i - θ ˉ 1) 2 n - 1$
加速踏板开度大于2/3的比例	$P 1$	$P 1 = M N × 100 %$
制动踏板开度均值	$θ ˉ 2$	$θ ˉ 2$ $= 1 n ∑ i = 1 n θ 2 i$
制动踏板开度标准差	$S θ 2$	$S θ 2 = ∑ i = 1 n (θ 2 i - θ ˉ 2) 2 n - 1$
加速度均值	$a ˉ$	$a ˉ = 1 n ∑ i = 1 n a i, a i > 0$
加速度标准差	$S a$	$S a = ∑ i = 1 n (a i - a ˉ) 2 n - 1$
减速度均值	$d ˉ$	$d ˉ = 1 n ∑ i = 1 n d i, d i < 0$
减速度标准差	$S d$	$S d = ∑ i = 1 n (d i - d ˉ) 2 n - 1$
冲击度绝对值均值	$J ˉ$	$J ˉ = 1 n ∑ i = 1 n J i$
冲击度绝对值标准差	$S J$	$S J = ∑ i = 1 n (J i - J ˉ) 2 n - 1$
单位里程能耗均值	$E ˉ E P K$	$E ˉ E P K$ $= 1 n ∑ i = 1 n E E P K i$

表1

表2

图2

图3

表3

成分矩阵"

$x l$	成分
$x l$	PC1	PC2	PC3	PC4
主成分命名	动力性指标	经济性指标	制动踏板开度	舒适性指标
$v ˉ$	0.203	0.68	0.294	-0.517
$S v$	0.681	-0.459	-0.048	-0.23
$a ˉ$	0.882	-0.174	0.182	0.065
$S a$	0.832	-0.167	-0.025	-0.003
$d ˉ$	-0.751	0.013	0.337	0.248
$S d$	0.678	0.074	-0.498	-0.206
$J ˉ$	0.153	0.779	-0.186	0.495
$S J$	0.656	0.288	-0.39	0.462
$E ˉ E P K$	0.338	-0.639	0.139	0.356
$θ ˉ 1$	0.794	0.244	0.471	-0.069
$S θ 1$	0.651	-0.03	0.197	0.277
$P 1$	0.802	0.185	0.47	0.052
$θ ˉ 2$	0.493	0.027	-0.439	-0.27

表3

图4

表4

图5

图6

表5

图7

图8

图9

图10

图11

图12

图13

表6

表7

图14

参考文献 26

1	Wang Wen-shuo, Xi Jun-qiang, Liu Chang, et al. Human-centered feed-forward control of a vehicle steering system based on a driver's path-following characteristics[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(6):1440-1453.
2	Martinez C M, Heucke M, Wang Fei-yue, et al. Driving style recognition for intelligent vehicle control and advanced driver assistance: a survey[J]. IEEE Transactions on Intelligent Transportation Systems, 2018, 19(3): 666-676.
3	Dib W, Chasse A, Moulin P, et al. Optimal energy management for an electric vehicle in eco-driving application[J]. Control Engineering Practice, 2014, 29(8): 299-307.
4	Javanmardi S, Bideaux E, Trégouët J, et al. Driving style modelling for eco-driving applications[J]. IFAC-Papers OnLine, 2017, 50(1): 13866-13871.
5	Butakov V, Ioannou P. Personalized driver/vehicle lane change models for ADAS[J]. IEEE Transactions on Vehicular Technology, 2015, 64(10): 4422-4431.
6	Higgs B, Abbas M. Segmentation and clustering of car-following behavior: recognition of driving patterns[J]. IEEE Transactions on Intelligent Transportation Systems, 2015, 16(1): 81-90.
7	Mohammadnazar A, Arvin R, Khattak A J. Classifying travelers' driving style using basic safety messages generated by connected vehicles: application of unsupervised machine learning[J]. Transportation Research Part C: Emerging Technologies, 2021, 122:102917.
8	Forster D, Inderka R B, Gauterin F. Data-driven identification of characteristic real-driving cycles based on k-means clustering and mixed-integer optimization[J]. IEEE Transactions on Vehicular Technology, 2019, 69(3): 2398-2410.
9	Wang W S, Xi J Q, Chong A, et al. Driving style classification using a semisupervised support vector machine[J]. IEEE Transactions on Human Machine Systems, 2017, 47(5): 650-660.
10	Han Wei, Wang Wen-Shuo, Li Xiao-Han, et al. Statistical-based approach for driving style recognition using bayesian probability with kernel density estimation[J]. IET Intelligent Transport Systems, 2018, 13(1): 22-30.
11	Wang Wen-Shuo, Xi Jun-Qiang, Zhao Ding. Driving style analysis using primitive driving patterns with bayesian nonparametric approaches[J]. IEEE Transactions on Intelligent Transportation Systems, 2018,20(8): 2986-2998.
12	严英. 纯电动公交客车加速踏板驾驶特性辅助优化策略研究[D].天津:天津大学机械工程学院, 2012.
	Yan Ying. Study on the driving assistant of accleration pedal for the pure electric bus[D]. Tianjin: School of Mechanical Engineering, Tianjin University, 2012.
13	叶玉.西方国家力推“生态驾驶”[J].环境与保护,2008,403(9):73-74.
	Ye Yu. Western countries push "eco-driving"[J]. Environmental Protection,2008,403(9):73-74.
14	朱嘉欣, 包雨恬, 黎朝. 数据离群值的检验及处理方法讨论[J]. 大学化学, 2018, 33(8): 58-65.
	Zhu Jia-xin, Bao Yu-tian, Li Zhao. Discussion on the method for testing and treating outliers[J]. University Chemistry, 2018, 33(8): 58-65.
15	Meenpal T, Goyal A, Meenpal A. Face recognition system based on principal components analysis and distance measures[J]. International Journal of Engineering Technologies IJET, 2021, 7(2): 15-19.
16	吴丽宁. 基于驾驶风格分类的卡车油耗预测[D].西安:长安大学信息工程学院,2020.
	Wu Li-ning. Truck fuel consumption prediction based on driving style classification[D]. Xi'an: School of Information Engineering, Chang'an Universty, 2020.
17	Johnstone L M, Lu A Y. On consistency and sparsity for principal components analysis in high dimensions[J]. Journal of the American Statistical Association, 2009, 104(486): 682-693.
18	Clayman C L, Srinivasan S M, Sangwan R S. K-means clustering and principal components analysis of microarray data of L1000 landmark genes[J]. Procedia Computer Science, 2020, 168:97-104.
19	Fu Rui, Liu Tong, Guo Ying-shi, et al. A case study in china to determine whether gps data and derivative indicator can be used to identify risky drivers[J]. Journal of Advanced Transportation, 2019, 2019:1-16.
20	Hidayati R, Zubair A, Pratama A H, et al. Analisis silhouette coefficient pada 6 perhitungan jarak k-means clustering[J]. Techno Com, 2021, 20(2): 186-197.
21	Yang Wen-kang, Du Qiao-ling, Cui Jian-chao, et al. Motion recognition based on sum of the squared errors distribution[J]. IEEE Access, 2021, 9: 37116-37130.
22	Lima S P, Cruz M. A genetic algorithm using calinski-harabasz index for automatic clustering problem[J]. Revista Brasileira de Computação Aplicada, 2020, 12(3): 97-106.
23	Wang Yi-mei, Liu Yong-qian, Li Li, et al. Short-term wind power forecasting based on clustering pre-calculated CFD method[J]. Energies, 2018, 11(4): No.854.
24	余涛. 基于SVM和BP神经网络的短时交通流预测与实现[D]. 南京:南京邮电大学现代邮政学院, 2018.
	Yu Tao. Prediction and implementation of short-term traffic based on the combination model of SVM and BP neural networks[D]. Nanjing: School of Modern Posts, Nanjing University of Posts and Telecommunications, 2018.
25	孙宝磊,孙暠,张朝能,等.基于BP神经网络的大气污染物浓度预测[J].环境科学学报,2017,37(5):1864-1871.
	Sun Bao-lei, Sun Hao, Zhang Chao-neng, et al. Forecast of air pollutant concentrations by BP neutual network[J]. Acta Scientiae Circumstantiae, 2017,37(5):1864-1871.
26	伍毅平. 生态驾驶行为特征甄别及反馈优化方法研究[D].北京:北京工业大学城市建设学部,2017.
	Wu Yi-ping. Research on eco-driving behavior characteristics identification and feedback optimization method[D]. Bejing: Faculty of Architecture, Civil and Transportation Engineering, Beijing University of Technology, 2017.

相关文章 15

[1]	马壮林,崔姗姗,胡大伟,王晋. 限行政策下传统小汽车出行者出行方式选择[J]. 吉林大学学报(工学版), 2023, 53(7): 1981-1993.
[2]	尹超英,陆颖,邵春福,马健霄,许得杰. 考虑空间自相关的建成环境对通勤方式选择的影响[J]. 吉林大学学报(工学版), 2023, 53(7): 1994-2000.
[3]	宋灿灿,荆迪菲,谢俊峰,康可心. 设置广告牌的高速公路平曲线路段驾驶行为分析[J]. 吉林大学学报(工学版), 2023, 53(5): 1345-1354.
[4]	潘恒彦,王永岗,李德林,陈俊先,宋杰,杨钰泉. 基于交通冲突的长纵坡路段追尾风险评估及预测[J]. 吉林大学学报(工学版), 2023, 53(5): 1355-1363.
[5]	康耀龙,冯丽露,张景安,曹素娥. 基于谱聚类的不确定数据集中快速离群点挖掘算法[J]. 吉林大学学报(工学版), 2023, 53(4): 1181-1186.
[6]	姚荣涵,徐文韬,郭伟伟. 基于因子长短期记忆的驾驶人接管行为及意图识别[J]. 吉林大学学报(工学版), 2023, 53(3): 758-771.
[7]	卢凯,徐广辉,叶志宏,林永杰. 考虑清空时间的双向队首绿波协调控制数解算法[J]. 吉林大学学报(工学版), 2023, 53(2): 421-429.
[8]	曲福恒,钱超越,杨勇,陆洋,宋剑飞,胡雅婷. 基于多球分裂的增量式k-means聚类算法[J]. 吉林大学学报(工学版), 2022, 52(6): 1434-1441.
[9]	康耀龙,冯丽露,张景安,陈富. 基于谱聚类的高维类别属性数据流离群点挖掘算法[J]. 吉林大学学报(工学版), 2022, 52(6): 1422-1427.
[10]	张鑫,张卫华. 快速路合流区主线不同交通状态下的安全性分析[J]. 吉林大学学报(工学版), 2022, 52(6): 1308-1314.
[11]	纪少波,李洋,李萌,苏士斌,马晓龙,何绍清,贾国瑞,程勇. 纯电动共享汽车驾驶行为对能耗的影响[J]. 吉林大学学报(工学版), 2022, 52(4): 754-763.
[12]	邢海燕,刘超,徐成,陈玉环,王松弘泽. 基于粒子群优化模糊C焊缝等级磁记忆定量识别模型[J]. 吉林大学学报(工学版), 2022, 52(3): 525-532.
[13]	廖汉超,单汨源. 基于模糊聚类最大树算法的高层建筑基坑工程风险识别方法[J]. 吉林大学学报(工学版), 2022, 52(12): 2892-2897.
[14]	曲大义,赵梓旭,贾彦峰,王韬,刘琼辉. 基于Lennard-Jones势的车辆跟驰动力学特性及模型[J]. 吉林大学学报(工学版), 2022, 52(11): 2549-2557.
[15]	董春娇,董黛悦,诸葛承祥,甄理. 电动自行车出行特性及骑行决策行为建模[J]. 吉林大学学报(工学版), 2022, 52(11): 2618-2625.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

方法		数值
取样足够度的Kaiser?Meyer?Olkin 度量		0.759
Bartlett 的球形度检验	近似卡方	4505.552
	df	78
	Sig.	0.000

成分	簇1	簇2	簇3
PC1	-0.371	0.578	-0.852
PC2	-0.793	0.079	1.422
PC3	0.260	-0.485	0.823
PC4	0.460	-0.278	-0.173

语义指标	驾驶行为特征参数	第1簇	第2簇	第3簇
经济性	单位里程能耗	最大	居中	最小
动力性	速度标准差	居中	最大	最小
	加速度均值	居中	最大	最小
	加速度标准差	居中	最大	最小
	减速度绝对值均值	居中	最大	最小
	减速度标准差	居中	最大	最小
	加速踏板开度均值	最小	最大	居中
	加速踏板开度标准差	居中	最大	最小
	加速踏板开度超过2/3的比例	最小	最大	居中
舒适性	冲击度绝对值均值	居中	最大	最小
舒适性	冲击度绝对值标准差	居中	最大	最小
考虑能耗的驾驶风格语义		该类驾驶行为表现出最差的驾驶经济性，对动力性和舒适性的追求不显著，定义其为一般型驾驶风格	该类驾驶行为表现出较差的驾驶经济性，追求动力性，行为表现最激进，舒适性也最差，定义其为耗能激进型驾驶风格	该类驾驶行为表现出最好的驾驶经济性，追求平稳的驾驶，看重舒适性，定义其为节能舒适型驾驶风格

实际风格	识别风格			识别率/%	平均识别率/%
实际风格	1	2	3	识别率/%	平均识别率/%
1	25	2	0	92.59	93.52
2	1	33	0	97.06
3	0	1	10	90.91

相关统计值	片段序号
相关统计值	1	2
速度均值/（km·h^-1）	28.92	25.54
速度标准差	14.19	12.71
加速踏板开度均值/%	74.17	47.7
加速踏板开度标准差	24.91	23.33
加速踏板行程超过2/3的比例/%	62.16	25.81
制动踏板开度均值/%	23.74	11.9
制动踏板开度标准差	3.48	9.57
加速度均值/（m·s^-2）	0.82	0.64
加速度标准差	0.55	0.49
减速度均值/（m·s^-2）	0.9	0.73
减速度标准差	0.48	0.45
单位里程能耗/（kW·h·km^-1）	2.06	0.45
总能耗/（kW·h）	0.58	0.13
行驶距离/m	280.4	282.4
行程时间/s	48	43.5
行驶时间/s	30	34.5
加速占比/%	48.33	27.54
减速占比/%	30	39.13
匀速占比/%	21.67	33.33