• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2023, Vol. 45 ›› Issue (07): 1236-1244.

• 图形与图像 • 上一篇    下一篇

基于卷积和Transformer的断路器动触头跟踪方法研究

崔克彬,崔叶微   

  1. (华北电力大学(保定)计算机系,河北 保定 071003)
  • 收稿日期:2022-02-16 修回日期:2022-04-01 接受日期:2023-07-25 出版日期:2023-07-25 发布日期:2023-07-11
  • 基金资助:
    河北省自然科学基金(F2018502080)

A circuit breaker moving contact tracking methods based on convolution and Transformer

CUI Ke-bin,CUI Ye-wei   

  1. (Department of Computer Science,North China Electric Power University,Baoding 071003,China)
  • Received:2022-02-16 Revised:2022-04-01 Accepted:2023-07-25 Online:2023-07-25 Published:2023-07-11

摘要: 测量断路器动触头运动特性有助于断路器运行状态的故障诊断。目前大部分测量方法为接触式测试,普遍存在安装不便、测量精度低的问题。为此,提出一种新的可以实现非接触式测量的算法。首先利用多尺度特征融合结构融合提取的多层深度特征,其次利用引入卷积操作的改进Transformer结构进行特征增强,最后通过预测头预测跟踪结果。实验结果表明,该跟踪算法相较于原算法,跟踪成功率提升了2.6%,精确度提升了13.9%,可以实现准确跟踪,进而得到断路器行程-时间曲线图,合理地反映了断路器操动机构的动作特性。

关键词: 断路器, 目标跟踪, Transformer, 多尺度特征融合, 卷积神经网络

Abstract: Measuring the motion characteristics of circuit breaker moving contacts can help diagnose the operating status of the circuit breaker. Currently, most measurement methods are "contact" testing methods, which generally have problems with inconvenient installation and low measurement accuracy. Therefore, a new model that can achieve non-contact measurement method is proposed. Firstly, the multi-scale feature fusion structure is used to fuse the extracted multi-layer depth features. Secondly, the improved Transformer structure with introduced convolution operation is used for feature enhancement. Finally, the prediction head is used to predict the tracking results. Experimental analysis shows that compared with the original algorithm, the tracking success rate of the tracking algorithm has increased by 2.6%, and the precision has increased by 13.9%. The model can achieve accurate tracking and obtain the circuit breaker stroke time curve, which can reasonably reflect the action char-acteristics of the circuit breaker operating mechanism. 

Key words: circuit breaker, target tracking, Transformer, multi-scale feature fusion, convolutional neural network