Please wait a minute...
工程设计学报  2024, Vol. 31 Issue (5): 565-574    DOI: 10.3785/j.issn.1006-754X.2024.03.203
机械设计理论与方法     
融合短时傅里叶变换和卷积神经网络的托辊故障诊断方法
谢苗1(),孟庆爽1(),李博2,卢进南1,李玉岐1,杨志勇1,3
1.辽宁工程技术大学 机械工程学院,辽宁 阜新 123000
2.辽宁工程技术大学 电气与控制工程学院,辽宁 葫芦岛 125000
3.新疆露天矿智能生产与管控重点实验室,新疆 昌吉 831100
Idler fault diagnosis method fusing short-time Fourier transform and convolutional neural network
Miao XIE1(),Qingshuang MENG1(),Bo LI2,Jinnan LU1,Yuqi LI1,Zhiyong YANG1,3
1.School of Mechanical Engineering, Liaoning Technical University, Fuxin 123000, China
2.Faculty of Electrical and Control Engineering, Liaoning Technical University, Huludao 125000, China
3.Xinjiang Open-pit Mine Intelligent Production and Control Key Laboratory, Changji 831100, China
 全文: PDF(5856 KB)   HTML
摘要:

托辊故障已成为带式输送机运行中的常见问题。若不能及时诊断托辊故障,则将严重制约带式输送机的安全运行。为了解决上述问题,基于某矿带式输送机中间段托辊的实际运行工况,提出了一种融合短时傅里叶变换(short-time Fourier transform, STFT)和卷积神经网络(convolutional neural network, CNN)的托辊故障诊断方法。首先,以分布式光纤为基础,对托辊在正常、轴承损坏及筒皮断裂工况下运行时的振动信号进行采集并作STFT处理,得到对应的时频图样本集,并将其分为训练集和测试集。然后,将训练集输入CNN模型以进行诊断模型训练,在训练过程中不断更新不同工况下托辊的运行状态特征。最后,将训练好的CNN模型应用于测试集,并输出托辊运行状态的识别结果。结果表明,所构建的CNN模型的识别准确率高达99.6%。基于所提出的故障诊断方法,在某矿上开展现场实验,以进一步验证CNN模型的识别准确率。实验结果表明,CNN模型对带式输送机中间段托辊的运行状态有较高的识别准确率,可达96.5%,与测试集上的识别准确率仅相差3.1个百分点,说明所提出的故障诊断方法具有一定的可靠性。后续可通过不断增加不同工况下托辊的运行数据来提高该故障诊断方法的鲁棒性,这可为煤矿企业有效诊断托辊故障提供有力的理论基础。

关键词: 托辊故障诊断分布式光纤短时傅里叶变换卷积神经网络    
Abstract:

Idler fault has become a common problem in the operation of belt conveyors. If the idler fault cannot be diagnosed in time, it will seriously restrict the safe operation of belt conveyors. To solve the above problems, based on the actual operating conditions of idlers in the middle section of a certain mining belt conveyor, an idler fault diagnosis method fusing short-time Fourier transform (STFT) and convolutional neural network (CNN) is proposed. Firstly, based on distributed optical fiber, the vibration signals of the idler operating under normal, bearing damage and cylinder skin fracture conditions were collected and processed by STFT to obtain corresponding time-frequency image sample set, and the sample set was divided into training set and testing set. Then, the training set was input into the CNN model for diagnostic model training, and the operating state characteristics of idlers under different working conditions were constantly updated during the training process. Finally, the trained CNN model was applied to the testing set, and the recognition result of the idler operating state was output. The results showed that the recognition accuracy of the constructed CNN model was as high as 99.6%. Based on the proposed fault diagnosis method, field experiment were carried out in a certain mine to further verify the recognition accuracy of the CNN model. The experimental results showed that the CNN model had a high recognition accuracy of 96.5% for the operating state of idlers in the middle section of the belt conveyor, which was 3.1 percentage points lower than the recognition accuracy on the testing set, indicating that the proposed fault diagnosis method had a certain reliability. Subsequently, the robustness of the fault diagnosis method can be improved by continuously increasing the operation data of idlers under different working conditions, which can provide a powerful theoretical basis for the effective diagnosis of idler faults in coal mine enterprises.

Key words: idler    fault diagnosis    distributed optical fiber    short-time Fourier transform    convolutional neural network
收稿日期: 2023-09-12 出版日期: 2024-10-30
CLC:  TH 222  
基金资助: 国家自然科学基金面上项目(51874158)
通讯作者: 孟庆爽     E-mail: xiemiao1121@126.com;2724127290@qq.com
作者简介: 谢 苗(1980—),女,教授,博士,从事综掘、综采成套装备研究,E-mail: xiemiao1121@126.comhttps://orcid. org/0000-0002-2980-0365
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
谢苗
孟庆爽
李博
卢进南
李玉岐
杨志勇

引用本文:

谢苗,孟庆爽,李博,卢进南,李玉岐,杨志勇. 融合短时傅里叶变换和卷积神经网络的托辊故障诊断方法[J]. 工程设计学报, 2024, 31(5): 565-574.

Miao XIE,Qingshuang MENG,Bo LI,Jinnan LU,Yuqi LI,Zhiyong YANG. Idler fault diagnosis method fusing short-time Fourier transform and convolutional neural network[J]. Chinese Journal of Engineering Design, 2024, 31(5): 565-574.

链接本文:

https://www.zjujournals.com/gcsjxb/CN/10.3785/j.issn.1006-754X.2024.03.203        https://www.zjujournals.com/gcsjxb/CN/Y2024/V31/I5/565

图1  不同运行状态的托辊
图2  托辊振动信号采集方法
图3  后向瑞利散射干涉效应的离散模型
图4  振动前后的瑞利散射时域曲线
图5  CNN的结构
图6  基于STFT和CNN的托辊故障诊断流程
图7  不同工况下托辊的振幅—时间曲线
图8  不同工况下托辊振动信号的时频图
图9  适用于托辊故障诊断的CNN模型
网络层特征图数量/张特征图大小/像素
输入层164×64
池化层1132×32
卷积层13232×32
池化层23216×16
卷积层26416×16
池化层3648×8
全连接层12 048
输出层11
表1  CNN的结构参数
图10  CNN模型在训练集和测试集上的识别准确率
图11  CNN模型在训练集和测试集上的损失值
图12  不同带速下CNN模型的识别准确率对比
图13  不同带速下新CNN模型的识别准确率对比
图14  带式输送机运行现场
图15  CNN模型的故障诊断结果与实际故障结果对比
图16  CNN模型对不同托辊运行状态的识别结果
1 齐庆杰, 王欢, 董子文, 等. 矿井胶带运输巷火灾蔓延规律的数值模拟研究[J]. 中国安全科学学报, 2016, 26(10): 36-41. doi:10.16265/j.cnki.issn1003-3033.2016.10.007
QI Q J, WANG H, DONG Z W, et al. Numerical simulation of belt conveyor fire spreading law in coal mine[J]. China Safety Science Journal, 2016, 26(10): 36-41.
doi: 10.16265/j.cnki.issn1003-3033.2016.10.007
2 李铬, 李春广, 梁睦, 等. 煤矿带式输送机事故分析及防护措施[J]. 中国安全科学学报, 2006, 16(3): 140-144, 148. doi:10.3969/j.issn.1003-3033.2006.03.027
LI G, LI C G, LIANG M, et al. Accident analysis of belt conveyor used in coal mine and its protective measures[J]. China Safety Science Journal, 2006, 16(3): 140-144, 148.
doi: 10.3969/j.issn.1003-3033.2006.03.027
3 赵学智, 叶邦彦, 陈统坚. 短时傅里叶变换的时频聚集性度量准则研究[J]. 振动、测试与诊断, 2017, 37(5): 948-956, 1065. doi:10.16450/j.cnki.issn.1004-6801.2017.05.015
ZHAO X Z, YE B Y, CHEN T J. Study on measure rule of time-frequency concentration of short time Fourier transform[J]. Journal of Vibration, Measurement & Diagnosis, 2017, 37(5): 948-956, 1065.
doi: 10.16450/j.cnki.issn.1004-6801.2017.05.015
4 张立智, 徐卫晓, 井陆阳, 等. 基于二维深度卷积网络的旋转机械故障诊断[J]. 机械强度, 2020, 42(5): 1039-1044.
ZHANG L Z, XU W X, JING L Y, et al. Rotating machinery fault diagnosis based on two-dimensional convolution neural network[J]. Journal of Mechanical Strength, 2020, 42(5): 1039-1044.
5 付忠广, 王诗云, 高玉才, 等. 基于Mobile-VIT的旋转机械故障诊断方法[J]. 汽轮机技术, 2023, 65(2): 119-121, 86.
FU Z G, WANG S Y, GAO Y C, et al. A fault diagnosis method of rotating machinery based on Mobile-VIT[J]. Turbine Technology, 2023, 65(2): 119-121, 86.
6 隆军, 吴金强. STFT和HHT在风力机轴承故障诊断中的应用[J]. 噪声与振动控制, 2013, 33(4): 219-222. doi:10.3969/j.issn.1006-1335.2013.04.045
LONG J, WU J Q. Application of short time Fourier transform and Hilbert-Huang transform in fault diagnosis of rolling bearings of windmill[J]. Noise and Vibration Control, 2013, 33(4): 219-222.
doi: 10.3969/j.issn.1006-1335.2013.04.045
7 李恒, 张氢, 秦仙蓉, 等. 基于短时傅里叶变换和卷积神经网络的轴承故障诊断方法[J]. 振动与冲击, 2018, 37(19): 124-131.
LI H, ZHANG Q, QIN X R, et al. Fault diagnosis method for rolling bearings based on short-time Fourier transform and convolution neural network[J]. Journal of Vibration and Shock, 2018, 37(19): 124-131.
8 金江涛, 许子非, 李春, 等. 基于深度学习与混沌特征融合的滚动轴承故障诊断[J]. 控制理论与应用, 2022, 39(1): 109-116. doi:10.7641/CTA.2021.10177
JIN J T, XU Z F, LI C, et al. Rolling bearing fault diagnosis based on deep learning and chaotic feature fusion[J]. Control Theory & Applications, 2022, 39(1): 109-116.
doi: 10.7641/CTA.2021.10177
9 刘慧斌, 李少波, 张安思, 等. 基于深度卷积神经网络的轴承多故障诊断研究[J]. 组合机床与自动化加工技术, 2020(5): 12-16. doi:10.1109/cac51589.2020.9326576
LIU H B, LI S B, ZHANG A S, et al. Multiple damage diagnosis of bearings based on deep convolutional neural network[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020(5): 12-16.
doi: 10.1109/cac51589.2020.9326576
10 JANSSENS O, SLAVKOVIKJ V, VERVISCH B, et al. Convolutional neural network based fault detection for rotating machinery[J]. Journal of Sound and Vibration, 2016, 377: 331-345.
11 董辛旻, 韩捷, 石来德, 等. 基于矢谱和L-M神经网络的旋转机械故障诊断研究[J]. 汽轮机技术, 2009, 51(5): 372-375.
DONG X M, HAN J, SHI L D, et al. Study on fault diagnosis for rotary machinery based on vector spectrum and L-M neural network[J]. Turbine Technology, 2009, 51(5): 372-375.
12 谢厚抗, 鲍久圣, 葛世荣, 等. 带式输送机承载托辊旋转阻力特性试验研究[J]. 煤炭学报, 2019, 44(): 731-736. doi:10.13225/j.cnki.jccs.2019.0902
XIE H K, BAO J S, GE S R, et al. Experimental research on rotational resistance characteristics of belt conveyor bearing idler[J]. Journal of China Coal Society, 2019, 44(): 731-736.
doi: 10.13225/j.cnki.jccs.2019.0902
13 李新华, 梁浩, 徐伟弘, 等. 常用分布式光纤传感器性能比较[J]. 光通信技术, 2007, 31(5): 14-18.
LI X H, LIANG H, XU W H, et al. Comparison of characteristics of commonly-used distributed optical fiber sensors[J]. Optical Communication Technology, 2007, 31(5): 14-18.
14 徐佳杰. 基于双树复小波与宽度学习的轴承故障诊断方法研究[D]. 包头: 内蒙古科技大学, 2020: 18-20.
XU J J. Research on diagnosis of bearing based on double tree complex wavelet and broad learning system [D]. Baotou: Inner Mongolia University of Science & Technology, 2020: 18-20.
15 KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 60(6): 84-90.
16 LECUN Y, BOSER B, DENKER J S, et al. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation, 1989, 1(4): 541-551.
17 袁铭阳. 基于深度学习的遥感影像典型目标检测算法研究[D]. 郑州: 战略支援部队信息工程大学, 2020: 17-21.
YUAN M Y. Research on typical target detection algorithm of remote sensing image based on deep learning[D]. Zhengzhou: Strategic Support Force Information Engineering University, 2020: 17-21.
18 储颜雨. 基于深度残差网络的细粒度图像分类研究[D]. 南京: 南京邮电大学, 2020: 24-27.
CHU Y Y. Research on fine-grained image classification based on deep residual network[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2020: 24-27.
19 李大柱, 牛江, 梁树林, 等. 基于多尺度时频图与卷积神经网络的车轮故障智能诊断[J]. 铁道科学与工程学报, 2023, 20(3): 1032-1043.
LI D Z, NIU J, LIANG S L, et al. Intelligent wheel fault diagnosis based on multi-scale time-frequency map and convolutional neural network[J]. Journal of Railway Science and Engineering, 2023, 20(3): 1032-1043.
[1] 沈剑雄,刘迎圆,王乐勤. 基于深度学习的翼型参数化建模方法[J]. 工程设计学报, 2024, 31(3): 292-300.
[2] 叶子汉,王中华,姜潮,吕新,张哲. 基于多判别器辅助分类器生成对抗网络的故障诊断方法研究[J]. 工程设计学报, 2024, 31(2): 137-150.
[3] 田立勇,唐瑞,于宁,陈洪月. 带式输送机更换托辊用皮带举升机构设计与应用[J]. 工程设计学报, 2023, 30(6): 667-677.
[4] 纪超,王亮,王孝敬,李小兵,曹雯. 基于MSSA-SVM的电缆隧道故障预警系统设计[J]. 工程设计学报, 2023, 30(1): 109-116.
[5] 吴国沛, 余银犬, 涂文兵. 永磁同步电机故障诊断研究综述[J]. 工程设计学报, 2021, 28(5): 548-558.
[6] 倪洪启, 金驰, 冯霏. 波纹补偿器故障诊断系统研制[J]. 工程设计学报, 2019, 26(3): 354-363.
[7] 马天兵, 王孝东, 杜菲, 王鑫泉. 基于GA-SVM的刚性罐道故障诊断[J]. 工程设计学报, 2019, 26(2): 170-176.
[8] 胡艺耀, 朱斌, 张伟, 何畏, 沈平生. 知识库构建工具软件的设计与实现[J]. 工程设计学报, 2018, 25(4): 367-373.
[9] 李晓豁,翁正洋,钱亚森,史尚伟,李 岩. 基于果蝇算法优化模糊RBF网络的液压破碎锤故障诊断[J]. 工程设计学报, 2015, 22(6): 540-545.
[10] 李玲玲, 景丽婷, 马东娟, 李志刚. 改进证据理论及其在电力系统故障诊断中的应用[J]. 工程设计学报, 2012, 19(6): 485-488.
[11] 韩海涛,马红光,韩琨,郑耿乐. 关于Volterra频域核辨识的多音激励信号设计[J]. 工程设计学报, 2012, 19(2): 123-127.
[12] 郝志勇, 刘伟, 夏玮, 闫闯. 基于BP神经网络的吸运风机故障诊断[J]. 工程设计学报, 2012, 19(1): 57-60.
[13] 韩海涛, 马红光, 李飞, 张家良. 基于系统输出频率响应函数的非线性系统研究[J]. 工程设计学报, 2011, 18(5): 373-376.
[14] 郭玉秀, 倪晓红, 王玉田. 基于混沌弱信号检测的轧机故障诊断研究[J]. 工程设计学报, 2011, 18(3): 218-221.
[15] 王成刚, 王学伟, 杨智勇, 曾儒伟. 测试性建模中故障模式概率更新算法研究[J]. 工程设计学报, 2009, 16(4): 252-255.