改进麻雀算法和Q-Learning优化集成学习轨道电路故障诊断

徐凯; 郑浩; 涂永超; 吴仕勋

doi:10.19713/j.cnki.43-1423/u.T20222268

您当前的位置：

首页 >

文章列表页 >

改进麻雀算法和Q-Learning优化集成学习轨道电路故障诊断

工程诊所•工匠之家 | 更新时间：2023-12-05

- 改进麻雀算法和Q-Learning优化集成学习轨道电路故障诊断
- Fault diagnosis of track circuit based on improved sparrow search algorithm and Q-Learning optimization for ensemble learning
- 铁道科学与工程学报 2023年20卷第11期页码：4426-4437
- 作者机构：
  
  1.重庆交通大学信息科学与工程学院，重庆 400074
  2.重庆市公共交通运营大数据工程技术研究中心，重庆 400074
- 作者简介：
  
  徐凯(1970—)，男，湖北宜昌人，教授，博士，从事机器学习、计算智能和列车控制研究；E-mail：xkxjxwx@hotmail.com
- 基金信息：
  
  四川省科技厅川渝合作重点研发项目(2020YFQ0057);重庆市自然科学基金资助项目(cstc2021jcyj-msxmX0017);重庆市教委科学技术研究项目(KJQN202000703)
- DOI：10.19713/j.cnki.43-1423/u.T20222268
  中图分类号：
扫描看全文
徐凯,郑浩,涂永超等.改进麻雀算法和Q-Learning优化集成学习轨道电路故障诊断[J].铁道科学与工程学报,2023,20(11):4426-4437.

XU Kai,ZHENG Hao,TU Yongchao,et al.Fault diagnosis of track circuit based on improved sparrow search algorithm and Q-Learning optimization for ensemble learning[J].Journal of Railway Science and Engineering,2023,20(11):4426-4437.
徐凯,郑浩,涂永超等.改进麻雀算法和Q-Learning优化集成学习轨道电路故障诊断[J].铁道科学与工程学报,2023,20(11):4426-4437. DOI： 10.19713/j.cnki.43-1423/u.T20222268.

XU Kai,ZHENG Hao,TU Yongchao,et al.Fault diagnosis of track circuit based on improved sparrow search algorithm and Q-Learning optimization for ensemble learning[J].Journal of Railway Science and Engineering,2023,20(11):4426-4437. DOI： 10.19713/j.cnki.43-1423/u.T20222268.

摘要

无绝缘轨道电路的故障具有复杂性与随机性，采用单一的模型进行故障诊断，其性能评价指标难以提高。而采用集成学习方式，则存在各基学习器结构、参数设计盲目，集成模型中各基学习器组合权重难以分配的问题。针对以上问题，提出一种改进麻雀算法和Q-Learning优化集成学习的轨道电路故障诊断新方法，该方法有机地将集成学习与计算智能和强化学习相结合，充分挖掘轨道电路故障特征，提高性能评价指标。首先，使用卷积神经网络、长短期记忆网络和多层感知器深度学习模型，以及支持向量机和随机森林传统机器学习模型，共同构成集成学习基学习器，解决单一学习模型的不足，不同基学习器的使用保证集成学习的多样性。从自动化机器学习角度出发，采用改进麻雀算法优化该集成学习模型的结构和参数，克服其结构和参数难以确定的问题。在此之上，引入强化学习Q-learning对集成模型中各基学习器组合权重进行优化，智能地确定集成学习各基学习器的组合权重。最后，将集成学习模型的预测结果与真实结果比较后得到误差，再采用BP神经网络对预测结果进行补偿修正，进一步提高轨道电路的故障诊断性能评价指标。仿真结果表明，利用所提方法进一步改善了轨道电路故障诊断的准确度、精确度、召回率和,F,1,值等性能评价指标。

Abstract

Due to the complexity and randomness of jointless track circuit faults, it is difficult to improve the performance evaluation index of fault diagnosis with a single model. There are several problems for the use of ensemble learning, such as the structure and parameters of each base learner are blind design. The weights of each base learner combination are difficult to be assigned in the ensemble model. In order to address these issues, a new fault diagnosis approach of track circuit based on the improved sparrow search algorithm and Q-Learning optimization for ensemble learning was proposed. The proposed method can organically combine ensemble learning with computational intelligence and reinforcement learning to fully exploit the fault characteristics of track circuit and improve the performance evaluation index. Firstly, the use of convolutional neural networks, long and short-term memory networks and multilayer perceptron deep learning models, as well as support vector machines and random forests traditional machine learning models, together constitute an ensemble learning base learner, which addresses the shortcomings of a single learning model and ensures the diversity of ensemble learning. From the perspective of automated machine learning, the improved sparrow algorithm is used to optimize the structure and parameters of ensemble learning model to overcome the problem that its structure and parameters are difficult to determine. On this basis, reinforcement learning Q-learning was introduced to optimize the combined weights of each base learner in the ensemble model. The combined weights of each base learner of ensemble learning were determined intelligently. Finally, the error was obtained by comparing the prediction results of the ensemble learning model with the real results. The BP neural network was used to compensate and correct the prediction results, which further improves the fault diagnosis performance evaluation index of track circuits. The simulation experiments show that the performance evaluation indexes such as accuracy, precision, recall and F1 value of track circuits fault diagnosis are further improved by using our proposed method.

关键词

无绝缘轨道电路故障诊断集成学习改进麻雀算法Q-learning误差修正

Keywords

jointless track circuitfault diagnosisensemble learningimproved sparrow search algorithmQ-learningerror correction

references

米根锁, 杨润霞, 梁利. 基于组合模型的轨道电路复杂故障诊断方法研究[J]. 铁道学报, 2014, 36(10): 65-69.

MI Gensuo, YANG Runxia, LIANG Li. Research on diagnosis method of complex fault diagnosis of track circuit based on combined model[J]. Journal of the China Railway Society, 2014, 36(10): 65-69.

张喜, 杜旭升, 刘朝英. 车站信号控制设备故障诊断专家系统的研究与实现[J]. 铁道学报, 2009, 31(3): 43-49.

ZHANG Xi, DU Xusheng, LIU Chaoying. Development of railway station signaling control equipment fault diagnosis expert system[J]. Journal of the China Railway Society, 2009, 31(3): 43-49.

LI Junwu, LI Guoning. Research on fault diagnosis of ZPW-2000K track circuit based on RS-BN algorithm[J]. Tehnicki Vjesnik-Technical Gazette, 2019, 26(4): 1091-1097.

陈星, 董昱. 基于模糊认知图模型的轨道电路故障诊断[J]. 铁道科学与工程学报, 2017, 14(9): 1983-1989.

CHEN Xing, DONG Yu. Track circuit fault diagnosis based on fuzzy cognitive map model[J]. Journal of Railway Science and Engineering, 2017, 14(9): 1983-1989.

郑云水, 牛行通, 康毅军. 蝙蝠算法优化模糊神经网络的25 Hz相敏轨道电路故障诊断研究[J]. 铁道学报, 2018, 40(12): 93-100.

ZHENG Yunshui, NIU Xingtong, KANG Yijun. Fault diagnosis research for 25 Hz phase sensitive track circuit based on bat algorithm to optimize fuzzy neural network [J]. Journal of the China Railway Society, 2018, 40(12): 93-100.

陈光武, 高亚丽, 焦相萌. 基于自适应变异SAPSO-LSSVM的轨道电路故障诊断[J]. 北京交通大学学报, 2021, 45(2): 1-7.

CHEN Guangwu, GAO Yali, JIAO Xiangmeng. Track circuit fault diagnosis based on adaptive mutation SAPSO-LSSVM[J]. Journal of Beijing Jiaotong University, 2021, 45(2): 1-7.

卢皎, 禹建丽, 黄春雷, 等. 基于卷积神经网络的ZPW-2000R轨道电路智能故障诊断方法[J]. 工业工程, 2021, 24(4): 127-133.

LU Jiao, YU Jianli, HUANG Chunlei, et al. A fault diagnosis method of ZPW-2000R track circuit based on convolutional neural network[J]. Industrial Engineering Journal, 2021, 24(4): 127-133.

DE BRUIN T, VERBERT K, BABUŠKA R. Railway track circuit fault diagnosis using recurrent neural networks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(3): 523-533.

谢旭旭, 戴胜华. 基于深度学习的无绝缘轨道电路故障诊断研究[J]. 铁道学报, 2020, 42(6):79-85.

XIE Xuxu, DAI Shenghua. Fault diagnosis of jointless track circuit based on deep learning[J]. Journal of the China Railway Society, 2020, 42(6): 79-85.

ZHENG Ziyuan, DAI Shenghua, XIE Xuxu. Research on fault detection for ZPW-2000A jointless track circuit based on deep belief network optimized by improved particle swarm optimization algorithm[J]. IEEE Access, 2020, 8:175981-175997.

刘一博, 赵林海. 基于CSM系统的JTC故障智能诊断方法[J]. 铁道学报, 2021, 43(11): 78-87.

LIU Yibo, ZHAO Linhai. Method of intelligent diagnosis of JTC faults based on CSM system[J]. Journal of the China Railway Society, 2021, 43(11): 78-87.

董煜, 董昱. 基于改进证据理论的轨道电路故障诊断方法[J]. 铁道科学与工程学报, 2019, 16(6): 1535-1542.

DONG Yu, DONG Yu. Fault diagnosis method of track circuit based on improved evidence theory[J]. Journal of Railway Science and Engineering, 2019, 16(6): 1535-1542.

董煜, 董昱. 基于方法组合的轨道电路故障诊断模型[J]. 铁道标准设计, 2019, 63(5):143-147.

DONG Yu, DONG Yu. Fault diagnosis model of track circuit based on method combination[J]. Railway Standard Design, 2019, 63(5): 143-147.

周志华.集成学习: 基础与算法[M]. 李楠, 译. 北京: 电子工业出版社, 2020.

ZHOU Zhihua. Ensemble methods: fundamentals and algorithms[M]. LI Nan, trans. Beijing: Publishing House of Electronics Industry,2020.

李新琴, 张鹏翔, 史天运, 等. 基于深度学习集成的高速铁路信号设备故障诊断方法[J]. 铁道学报, 2020, 42(12): 97-105.

LI Xinqin, ZHANG Pengxiang, SHI Tianyun, et al. Research on fault diagnosis method for high-speed railway signal equipment based on deep learning integration[J]. Journal of the China Railway Society, 2020, 42(12):97-105.

焦李成, 刘若辰, 慕彩红. 简明人工智能[M]. 西安: 西安电子科技大学出版社, 2019.

JIAO Licheng, LIU Ruochen, MU Caihong. Concise artificial intelligence[M]. Xi’an: Xidian University Press, 2019.

赵会兵, 陈建译,孙上鹏. 列车运行控制系统故障诊断方法[Ｍ]. 北京:中国铁道出版社, 2015.

ZHAO Huibing, CHEN Jianyi, SUN Shangpeng. Fault diagnosis method of train operation control system[M]. Beijing: China Railway Press, 2015.

薛建凯. 一种新型的群智能优化技术的研究与应用: 麻雀搜索算法[D]. 上海: 东华大学, 2020.

XUE Jiankai. Research and application of a new swarm intelligence optimization technology-sparrow search algorithm[D].Shanghai: Donghua University, 2020.

付华, 刘昊. 多策略融合的改进麻雀搜索算法及其应用[J]. 控制与决策, 2022, 37(1): 87-96.

FU Hua, LIU Hao. Improved sparrow search algorithm with multi-strategy integration and its application[J]. Control and Decision, 2022, 37(1): 87-96.

龙文, 伍铁斌, 唐明珠, 等. 基于透镜成像学习策略的灰狼优化算法[J]. 自动化学报, 2020, 46(10): 2148-2164.

LONG Wen, WU Tiebin, TANG Mingzhu, et al. Grey wolf optimizer algorithm based on lens imaging learning strategy[J]. Acta Automatica Sinica, 2020, 46(10): 2148-2164.

RAHNAMAYAN S, TIZHOOSH H R, SALAMA M M A. Opposition-based differential evolution[J]. IEEE Transactions on Evolutionary Computation, 2008, 12(1): 64-79.

张达敏, 陈忠云, 辛梓芸, 等. 基于疯狂自适应的樽海鞘群算法[J]. 控制与决策, 2020, 35(9): 2112-2120.

ZHANG Damin, CHEN Zhongyun, XIN Ziyun, et al. Salp swarm algorithm based on craziness and adaptive[J]. Control and Decision, 2020, 35(9): 2112-2120.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于贝叶斯元学习的小样本转辙机故障诊断

改进CEEMD-SVM的轴承故障识别方法及其应用

基于深度网络的轨道电路暂态特征多补偿电容故障定位

基于邻相电流的逆变器功率管与电流传感器故障诊断方法

自适应频域字典的机车轮对轴承稀疏诊断方法