基于深度学习的弓网异常状态在线检测研究

周兆安; 李树枝

doi:10.13890/j.issn.1000-128X.2022.05.020

您当前的位置：

首页 >

文章列表页 >

基于深度学习的弓网异常状态在线检测研究

铁道电气化 | 更新时间：2024-08-02

- 基于深度学习的弓网异常状态在线检测研究
- Research on online inspection of pantograph and catenary based on deep learning
- 机车电传动 2022年第5期页码：135-143
- 作者机构：
  
  中车青岛四方机车车辆股份有限公司，山东青岛　266111
- 作者简介：
  
  周兆安（1973—），男，工程师，主要从事动车组检修运用方面的研究；E-mail: zhouzhaoan@cqsf.com
- 基金信息：
- DOI：10.13890/j.issn.1000-128X.2022.05.020
  中图分类号： U225;O436
- 纸质出版日期：2022-09-10，
  
  收稿日期：2022-03-31，
  
  修回日期：2022-08-15，
- 稿件说明：
扫描看全文
周兆安, 李树枝. 基于深度学习的弓网异常状态在线检测研究[J]. 机车电传动, 2022,(5):135-143.

ZHOU Zhaoan, LI Shuzhi. Research on online inspection of pantograph and catenary based on deep learning[J]. Electric drive for locomotives, 2022,(5):135-143.
周兆安, 李树枝. 基于深度学习的弓网异常状态在线检测研究[J]. 机车电传动, 2022,(5):135-143. DOI： 10.13890/j.issn.1000-128X.2022.05.020.

ZHOU Zhaoan, LI Shuzhi. Research on online inspection of pantograph and catenary based on deep learning[J]. Electric drive for locomotives, 2022,(5):135-143. DOI： 10.13890/j.issn.1000-128X.2022.05.020.

摘要

弓网系统作为电力机车的关键供能系统，其运行状态直接决定了电力机车的受流质量，影响列车运行的安全和效率。为了解决传统弓网检测方法效率低、实时性差等问题，文章设计了一种基于深度学习的弓网状态在线检测系统方案，采用英伟达（NVIDIA）的Xavier SoC模块进行图像处理，通过YOLO v4实现弓网检测并添加自适应图像增强模块，优化前后目标检测的mAP分别为0.950和0.961（IOU阈值为0.5）；实现基于ViT轻量级注意力模型的吊弦状态分类，平均准确率为97.69%；使用NVIDIA的推理加速器TensorRT加速后，检测模型和分类模型的推理时间分别为31.0 ms和2.2 ms。系统具有较高的鲁棒性与实用性，可为后续弓网异常在线检测功能拓展提供理论依据和设计参考。

Abstract

The pantograph-catenary system is a key part of power supply for electric locomotive

and its operating status determines the current receiving quality of electric locomotive

as well as safety and efficiency of trains. In order to solve the problems of traditional method for inspection of pantograph-catenary status

such as low efficiency and poor performance in real time inspection

this paper designed an online pantograph-catenary status inspection system solution based deep learning. The solution adopted NIDIA Xavier SoC module to perform image processing

and realizes pantograph-catenary inspection by YOLO v4 and adaptive image enhancement module was also added. The mAP of inspection target before and after optimization was 0.950 and 0.961 (with the IOU threshold of 0.5) respectively. Classification of catenary dropper status based on ViT lightweight class attention model was realized at an average accuracy rate of 97.69%. After acceleration by using NVIDIA TensorRT accelerator

the inference time of inspection model and classification model were 31.0 ms and 2.2 ms respectively. The system has high robustness and practicability

which can provide theory basis and design reference for online inspection function of pantograph-catenary abnormality in the future.

关键词

弓网状态检测接触网吊弦深度学习TensorRTViTYOLO v4

Keywords

pantograph-catenary status inspectioncatenarydropperdeep learningTensorRTViTYOLO v4

references

熊嘉阳, 沈志云. 中国高速铁路的崛起和今后的发展[J]. 交通运输工程学报, 2021, 21(5): 6-29.

XIONG Jiayang, SHEN Zhiyun. Rise and future development of Chinese high-speed railway[J]. Journal of Traffic and Transportation Engineering, 2021, 21(5): 6-29.

ZHANG Dongkai, GAO Shibin, YU Long, et al. A robust pantograph－catenary interaction condition monitoring method based on deep convolutional network[J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69(5): 1920-1929.

AYDIN I, KARAKOSE M, AKIN E. Anomaly detection using a modified kernel-based tracking in the pantograph－catenary system[J]. Expert Systems with Applications, 2015, 42(2): 938-948.

李宏科. 车载接触网运行状态检测装置(3C)的应用[J]. 通信电源技术, 2018, 35(6): 113-114.

LI Hongke. Application of vehicle mounted catenary running state detecting device (3C)[J]. Telecom Power Technology, 2018, 35(6): 113-114.

王同军. 高速铁路弓网检测监测体系研究[J]. 中国铁路, 2021(3): 1-7.

WANG Tongjun. Research on HSR pantograph-OCS inspection and monitoring system[J]. China Railway, 2021(3): 1-7.

张润宝, 杨志鹏. 接触网运行状态检测监测系统研究与实践[J]. 中国铁路, 2019(9): 64-70.

ZHANG Runbao, YANG Zhipeng. Research and practice of operation state inspection and monitoring system of overhead contact line system[J]. China Railway, 2019(9): 64-70.

连苏宁, 李轶, 姜彬, 等. 城市轨道交通接触网检测系统发展现状及趋势[J]. 科技成果管理与研究, 2019(1): 43-46.

LIAN Suning, LI Yi, JIANG Bin, et al. Development status and trend of catenary detection system in urban rail transit[J]. Management and Research on Scientific & Technological Achievements, 2019(1): 43-46.

ZHOU Ning, YANG Wenjie, LIU Jiurui, et al. Investigation on monitoring system for pantograph and catenary based on condition-based recognition of pantograph[J]. Shock and Vibration, 2019, 2019: 3839191.

BRUNI S, BUCCA G, COLLINA A, et al. Dynamics of the pantograph-catenary system[M]//Iwnicki S, Spiryagin M, COLE C, et al. Handbook of Railway Vehicle Dynamics. 2nd ed. Boca Raton: CRC Press, 2019: 37.

贺文, 王俊平, 张敏, 等. 车载接触网运行状态检测装置系统设计及应用[J]. 机车电传动, 2020(1): 144-148.

HE Wen, WANG Junping, ZHANG Min, et al. System design and application of on-board overhead catenary monitoring device[J]. Electric Drive for Locomotives, 2020(1): 144-148.

GAO Shibin. Automatic detection and monitoring system of pantograph－catenary in China’s high-speed railways[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-12.

张冰. 基于可控紫外光的弓网燃弧检测装置定标系统研究[J]. 铁道机车车辆, 2021, 41(5): 76-82.

ZHANG Bing. Study on calibration system of arcing detection device of pantograph and catenary based on controllable ultraviolet light[J]. Railway Locomotive & Car, 2021, 41(5): 76-82.

齐冬莲, 钱佳莹, 闫云凤, 等. 一种基于RefineDet网络和霍夫变换的高速铁路接触网吊弦状态多尺度检测方法[J]. 电子与信息学报, 2021, 43(7): 2014-2022.

QI Donglian, QIAN Jiaying, YAN Yunfeng, et al. A multi-scale detection method for dropper states in high-speed railway contact network based on RefineDet network and Hough transform[J]. Journal of Electronics & Information Technology, 2021, 43(7): 2014-2022.

胡碟, 金炜东, 唐鹏. 基于双尺度特征的铁路接触网吊弦检测[J]. 电工技术, 2020(15): 64-68.

HU Die, JIN Weidong, TANG Peng. Detection of railway dropper based on double scale features[J]. Electric Engineering, 2020(15): 64-68.

LIU Wenyu, REN Gaofeng, YU Runsheng, et al. Image-adaptive YOLO for object detection in adverse weather conditions[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(2): 1792-1800.

HE Kaiming, SUN Jian, TANG Xiaoou. Single image haze removal using dark channel prior[C]//IEEE. 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami: IEEE, 2009: 1956-1963.

HU Yuanming, HE Hao, XU Chenxi, et al. Exposure: a white-box photo post-processing framework[J]. ACM Transactions on Graphics, 2018, 37(2): 26.

DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16<math id="M44"><mo>×</mo></math>https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=63326452&type=2.28600001https://html.publish.founderss.cn/rc-pub/api/common/picture?pictureId=63326455&type=1.4393334416 words: transformers for image recognition at scale[C]//ICLR. International Conference on Learning Representations. Vienna: ICLR, 2021: 1-21.

VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Curran Associates Inc. Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach California: Curran Associates Inc., 2017: 6000-6010.

HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//IEEE. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 770-778.

BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLO v4: optimal speed and accuracy of object detection[DB/OL]. (2020-04-23) [2022-04-12]. https://arxiv.org/abs/2004.10934https://arxiv.org/abs/2004.10934.

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于深度卷积神经网络的铁路接触网鸟窝检测方法研究

基于三维点云与图像融合的轨道交通场景行人检测方法

基于低秩分解的YOLO轻量化目标检测模型

轨道交通车载端到端语音合成

基于改进yolov4-tiny的地铁受电弓燃弧检测方法