吉林大学学报(工学版) ›› 2021, Vol. 51 ›› Issue (5): 1692-1699.doi: 10.13229/j.cnki.jdxbgxb20200378

• 交通运输工程·土木工程 • 上一篇    

基于车联网信息的公交客车行驶工况数据处理方法

曾小华(),宋美洁,宋大凤(),王越   

  1. 吉林大学 汽车仿真与控制国家重点实验室,长春 130022
  • 收稿日期:2020-05-30 出版日期:2021-09-01 发布日期:2021-09-16
  • 通讯作者: 宋大凤 E-mail:13504422161@126.com;songdf@126.com
  • 作者简介:曾小华(1977-),男,教授,博士生导师.研究方向:新能源汽车关键技术.E-mail:13504422161@126.com
  • 基金资助:
    吉林省自然基金项目(YDZJ202101ZYTS159);吉林省科技发展计划智能制造重大科技专项项目(20200501010GX)

Data processing method of bus driving cycle based on vehicular network information

Xiao-hua ZENG(),Mei-jie SONG,Da-feng SONG(),Yue WANG   

  1. State Key Laboratory of Automotive Simulation and Control,Jilin University,Changchun 130022,China
  • Received:2020-05-30 Online:2021-09-01 Published:2021-09-16
  • Contact: Da-feng SONG E-mail:13504422161@126.com;songdf@126.com

摘要:

受技术、成本和环境等因素的影响,当前车联网平台数据在传输过程中不可避免地存在缺失和数据噪声问题。针对车联网数据传输优化需进一步改进的现实问题,基于某商用客车企业车联网平台,获取固定线路运行的公交客车行驶工况数据,分析研究车联网平台数据特点及其存在的问题,采用插补缺失与神经网络集成方法对行驶工况缺失数据进行估计,并基于小波变换对行驶工况噪声数据进行清洗。最后,给出了行驶工况数据处理方法的评价指标,合理说明了评价本文数据处理方法的效果。结果表明:本文提出的缺失数据轨迹方法,可较好地估计还原车辆真实工况的变化情况,基于小波变换的方法对噪声数据进行了有效处理,两者误差均在合理范围内。本文开展的研究内容可为车联网领域的数据传输优化改进、海量高并发数据的处理方法提供实际参考及借鉴。

关键词: 车辆工程, 公交客车, 车联网信息, 行驶工况, 数据缺失, 数据噪声

Abstract:

Under the influence of technology, cost, environment and other adverse factors, data missing and data noise problems inevitably exist in the data transmission process of the vehicle networking platform. In view of the practical problems that need to be further improved in the data transmission optimization of the Internet of vehicles, based on the vehicle networking platform of a commercial bus enterprise, this paper obtains the driving cycle data of the bus running on the fixed line. According to the characteristics and problems of the data of the vehicle networking platform, the missing data of driving cycle is estimated by the integration method of interpolation and neural network, and the noise data of driving cycle is cleaned based on wavelet transform. Finally, the evaluation index of the data processing method of driving cycle is given, and the effect of the data processing method is explained reasonably. The results show that the missing data track method can better estimate and restore the change of the real working condition of the vehicle. The method based on wavelet transform effectively processes the noise data, and the errors of both methods are within a reasonable range. The research content of this paper can provide practical reference and reference for the optimization of data transmission and the processing methods of massive and high concurrent data in the field of Internet of vehicles.

Key words: vehicle engineering, urban bus, vehicle networking information, driving cycle, data missing, data noise

中图分类号: 

  • U469.1

图1

7天行驶工况信息-车速变化"

图2

7天行驶信息-主电机转速的变化"

表1

车联网平台下行驶工况数据缺失情况"

车联网平台时间时间/s车速/ (km·h-1主电机转速/ (r·min-1
????
2018/12/15 6:33:05375271827
2018/12/15 6:33:15385322132
2018/12/15 6:33:27397252132
2018/12/15 6:33:35405312071
2018/12/15 6:33:50420291920
2018/12/15 6:34:02432291949
????

图3

车联网平台行驶工况数据缺失估计方法"

图4

车联网平台行驶工况数据缺失估计结果对比"

图5

不同缺失时间间隔下缺失数据估计"

图6

不同尺度小波分解下速度和加速度滤波情况"

图7

数据处理后的行驶工况曲线"

表2

行驶工况数据误差评价指标对比"

误差评价指标数值
MAE/(m·s-10.42
RMSE/(m·s-10.33
MRE0.27

表3

行驶工况特征参数指标对比"

工况特征参数指标真实工况数据处理后工况误差
平均车速/(km·h-130.330.20.1
平均巡航车速/(km·h-119.619.30.3
最高车速/(km·h-158.258.00.2
平均加速度/(m·s-20.320.300.02
最大加速度/(m·s-21.801.750.05
平均减速度/(m·s-2-0.31-0.290.02
怠速比例/%10.310.00.3
加速比例/%46.446.20.2
均速比例/%18.518.40.1
1 Zhang D G, Niu H L, Liu S, et al. Novel positioning service computing method for WSN[J]. Wireless Personal Communications, 2017, 92(4): 1747-1769.
2 Zhang D G, Song X D, Wang X, et al.New agent-based proactive migration method and system for big data environment (BDE)[J]. Engineering Computations, 2015, 32(8): 2443-2466.
3 张网娟, 许国艳, 李敏佳, 等. 基于卷积神经网络的缺失数据填充方法[J]. 微电子学与计算机, 2019, 36(3): 48-52.
Zhang Wang-juan, Xu Guo-yan, Li Min-jia, et al. Missing data imputation approach based on convolutional neural network[J]. Microelectronics & Computer, 2019, 36(3): 48-52.
4 高雨晨, 唐耀庚. 核电厂环境辐射监测传感器网络中缺失值的粒子群算法-最小二乘支持向量机估计算法[J]. 核电子学与探测技术, 2014, 34(12): 1508-1513.
Gao Yu-chen, Tang Yao-geng. PSO-LSSVM based missing data imputation algorithm for wireless sensor network for nuclear power plants' environmental radiation monitor[J]. Nuclear Electronics & Detection Technology, 2014, 34(12): 1508-1513.
5 王红宇, 游敏娟, 李琪, 等. 一种中值滤波图像去噪的改进算法[J]. 中国科技信息, 2019(1): 84-85.
Wang Hong-yu, You Min-juan, Li Qi, et al. An improved image denoising algorithm based on median filter[J]. China Science and Technology Information, 2019(1): 84-85.
6 伍灵杰. 数据采集系统中数字滤波算法的研究[D]. 北京: 北京林业大学信息学院, 2010.
Wu Ling-jie. Study on digital filter algorithms of data acquisition system[D]. Beijing: College of Information Science & Technology, Beijing Forestry University, 2010.
7 王小旭, 赵琳, 夏全喜, 等. 噪声相关条件下unscented卡尔曼滤波器设计[J]. 控制理论与应用, 2010, 27(10): 1362-1368.
Wang Xiao-xu, Zhao Lin, Xia Quan-xi, et al. Design of unscented Kalman filter with correlative noises[J]. Control Theory & Applications, 2010, 27(10): 1362-1368.
8 李昱辰, 李战明. 噪声未知情况下的自适应无迹粒子滤波算法[J]. 吉林大学学报: 工学版, 2013, 43(4): 1139-1145.
Li Yu-chen, Li Zhan-ming. Adaptive unscented particle filter algorithm under unknown noise[J]. Journal of Jilin University(Engineering and Technology Edition), 2013, 43(4): 1139-1145.
9 姜平, 石琴, 陈无畏, 等. 基于小波分析的城市道路行驶工况构建的研究[J]. 汽车工程, 2011, 33(1): 70-73.
Jiang Ping, Shi Qin, Chen Wu-wei, et al. A research on the construction of city road driving cycle based on wavelet analysis[J]. Automotive Engineering, 2011, 33(1): 70-73.
10 丁一峰, 李君, 盖洪超, 等. 小波变换在行驶工况构建用车速数据处理中的应用[J]. 科学技术与工程, 2017, 17(28): 274-279.
Ding Yi-feng, Li Jun, Gai Hong-chao, et al. Application of vehicle speed data processing in driving cycle construction with wavelet transform[J]. Science Technology and Engineering, 2017, 17(28): 274-279.
11 王文璇, 朱晓东, 吴兵. 基于动态时间弯曲距离的快速路入口匝道周围路径分级方法[J]. 吉林大学学报: 工学版, 2020, 50(3): 970-979.
Wang Wen-xuan, Zhu Xiao-dong, Wu Bing, et al. Utilizing dynamic time warping distance in classification of routes around expressway on-ramp[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(3): 970-979.
12 张健南. 缺失数据下极限学习机改进算法及其应用[D]. 北京: 清华大学自动化系, 2014.
Zhang Jian-nan. Improved algorithms based on extreme learning machine for handing with missing data and application[D]. Beijing: College of Automation, Tsinghua University, 2014.
13 Liu J, Chen Y, Zhan J, et al. An on-line energy management strategy based on trip condition prediction for commuter plug-in hybrid electric vehicles[J]. IEEE Transactions on Vehicular Technology, 2018, 67(5): 3767-3781.
[1] 罗勇,韦永恒,黄欢,肖人杰,任淋,崔环宇. 驾驶员意图识别的P2.5插混构型双离合器起步控制[J]. 吉林大学学报(工学版), 2021, 51(5): 1575-1582.
[2] 马超,高云凯,刘哲,段月星,田林雳. 骨架式车身多材料及梁截面形状和尺寸优化[J]. 吉林大学学报(工学版), 2021, 51(5): 1583-1592.
[3] 兰凤崇,李继文,陈吉清. 面向动态场景复合深度学习与并行计算的DG-SLAM算法[J]. 吉林大学学报(工学版), 2021, 51(4): 1437-1446.
[4] 杨建,夏琦,周海超,王国林. 修正胎体弦轮廓载重子午线轮胎的降噪机理[J]. 吉林大学学报(工学版), 2021, 51(4): 1198-1203.
[5] 龙江启,向锦涛,俞平,王骏骋. 适用于非线性主动悬架滑模控制的线性干扰观测器[J]. 吉林大学学报(工学版), 2021, 51(4): 1230-1240.
[6] 陈鑫,于贵申,张彪,潘凯旋,杨立飞. 搅拌摩擦点焊接头拉伸-剪切行为的等效建模[J]. 吉林大学学报(工学版), 2021, 51(4): 1190-1197.
[7] 宋大凤,杨丽丽,曾小华,王星琦,梁伟智,杨南南. 基于行驶工况合成的混合动力汽车电池寿命优化[J]. 吉林大学学报(工学版), 2021, 51(3): 781-791.
[8] 张家旭,王欣志,赵健,施正堂. 汽车高速换道避让路径规划及离散滑模跟踪控制[J]. 吉林大学学报(工学版), 2021, 51(3): 1081-1090.
[9] 何仁,赵晓聪,杨奕彬,王建强. 基于驾驶人风险响应机制的人机共驾模型[J]. 吉林大学学报(工学版), 2021, 51(3): 799-809.
[10] 王波,何洋扬,聂冰冰,许述财,张金换. 底部爆炸条件下车内乘员的腹部损伤[J]. 吉林大学学报(工学版), 2021, 51(3): 792-798.
[11] 庄蔚敏,王鹏跃,高瑞娟,解东旋. 温热成形对AA5754铝合金静态力学性能的影响[J]. 吉林大学学报(工学版), 2021, 51(3): 847-854.
[12] 宋强,孙丹婷,章伟. 纯电动车机械式自动变速器换挡非线性建模及控制[J]. 吉林大学学报(工学版), 2021, 51(3): 810-819.
[13] 李寿涛,王蕊,徐靖淳,王德军,田彦涛,于丁力. 一种基于模型预测复合控制的车辆避碰控制方法[J]. 吉林大学学报(工学版), 2021, 51(2): 738-746.
[14] 王伟达,武燕杰,史家磊,李亮. 基于驾驶员意图识别的电子液压制动助力系统控制策略[J]. 吉林大学学报(工学版), 2021, 51(2): 406-413.
[15] 陈国迎,姚军,王鹏,夏其坤. 适用于后轮轮毂驱动车辆的稳定性控制策略[J]. 吉林大学学报(工学版), 2021, 51(2): 397-405.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!