基于虚实融合数据的悬臂式掘进机截割部故障预警技术研究

张旭辉; 白琳娜; 杨红强

doi:10.13272/j.issn.1671-251x.2023050063

基于虚实融合数据的悬臂式掘进机截割部故障预警技术研究

doi: 10.13272/j.issn.1671-251x.2023050063

张旭辉^{1, 2,},
白琳娜¹,
杨红强¹

1.
西安科技大学机械工程学院, 陕西西安　710054
2.
陕西省矿山机电装备智能监测重点实验室, 陕西西安　710054

基金项目: 国家自然科学基金项目（52104166）；陕煤联合基金项目(2021JLM-03)。

详细信息

作者简介:
张旭辉（ 1972—），男，陕西凤翔人，教授，博士，研究方向为煤矿机电设备智能检测与控制， E-mail： zhangxh@xust.edu.cn

中图分类号: TD421
计量
- 文章访问数: 1103
- HTML全文浏览量: 57
- PDF下载量: 43
- 被引次数: 0
出版历程
- 收稿日期: 2023-05-18
- 修回日期: 2023-07-26
- 网络出版日期: 2023-09-04

Research on fault warning technology for cutting part of cantilever roadheader based on virtual and real fusion data

1.
College of Mechanical Engineering, Xi'an University of Science and Technology, Xi'an 710054, China
2.
Shaanxi Key Laboratoty of Mine Electromechanical Equipment Intelligenct Monitoring, Xi'an 710054, China

摘要

摘要: 目前悬臂式掘进机截割部故障预警技术依赖于传统的数据采集方法，在掘进机截割部实际运行过程中存在信号获取困难、噪声较多等问题，导致掘进机截割部故障预测预警能力受到限制。针对上述问题，提出一种基于虚实融合数据的悬臂式掘进机截割部故障预警方法。对悬臂式掘进机截割部进行三维实体建模，利用机械系统动力学自动分析软件(ADAMS)获取截割部机械系统虚拟数据，构建其动力学仿真模型以获取虚拟数据，并采用余弦相似度函数表征其与真实数据的相似度，验证虚拟数据的可信度。将虚拟、真实数据分别采用贝叶斯估计与自适应互补加权融合方法进行相似关联与互补关联融合，获得虚实融合数据。针对传统自组织映射（SOM）神经网络学习效率易受学习速率的影响问题，建立了基于改进SOM神经网络的故障预警模型，引入关于时间的单调递减函数对SOM神经网络进行训练，在保证学习速率的同时，兼顾模型的稳定性。将融合数据输入基于SOM神经网络的故障预警模型以确定获胜神经元并进行权值调整，计算真实数据与获胜神经元间的距离并进行权值调整，进而实现故障预警。实验结果表明，改进SOM神经网络的平均运行效率可提高35.84%；基于虚实融合数据的悬臂式掘进机截割部故障预警方法可成功实现单一故障和复合故障的类型预测，其预测准确率达83.33%。
- 悬臂式掘进机 /
- 截割部 /
- 故障预警 /
- 故障预测 /
- 虚实数据融合 /
- 自组织映射神经网络
Abstract: Currently, the fault warning technology for the cutting part of cantilever roadheader relies on traditional data collection methods. In the operation process of the cutting part of the roadheader, problems such as difficulty in obtaining signals and high noise limit the capability to predict and warn faults in the cutting part of the roadheader. In order to solve the above problems, a fault warning method for the cutting part of cantilever roadheader based on virtual and real fusion data is proposed. The method performs three-dimensional solid modeling of the cutting section of a cantilever roadheader. It uses the automatic dynamic analysis of mechanical systems (ADAMS) to obtain virtual data of the cutting section's mechanical system, constructs its dynamic simulation model to obtain virtual data. The method uses the cosine similarity function to characterize its similarity with real data to verify the credibility of the virtual data. The method uses Bayesian estimation and adaptive complementary weighted fusion methods to perform similarity association and complementary association fusion on virtual and real data, respectively, to obtain virtual and real fusion data. In response to the problem that the learning efficiency of traditional self-organizing mapping (SOM) neural networks is easily affected by the learning rate, a fault warning model based on an improved SOM neural network is established. A monotonic decreasing function about time is introduced to train the SOM neural network, ensuring both the learning rate and the stability of the model. The method inputs the fused data into the fault warning model based on SOM neural network to determine the winning neuron and adjust its weight. The method calculates the distance between the real data and the winning neuron and adjusts its weight to achieve fault warning. The experimental results show that the average operating efficiency of the improved SOM neural network can be improved by 35.84%. The fault warning method for the cutting part of a cantilever roadheader based on virtual and real fusion data can successfully predict the types of single and composite faults, with a prediction accuracy of 83.33%.
- cantilever roadheader /
- cutting part /
- fault warning /
- fault prediction /
- virtual and real data fusion /
- self-organizing mapping neural network

HTML全文

图 1 悬臂式掘进机截割部故障预警总体方案

Figure 1. Overall scheme of fault early warning for cutting part of cantilever roadheader

下载: 全尺寸图片幻灯片

图 2 截割部行星减速器动力学仿真模型

Figure 2. Dynamics simulation model of planetary reducer in cutting part

下载: 全尺寸图片幻灯片

图 3 正常状态下行星减速器传动轴（输入端）幅值

Figure 3. Amplitude diagram of planetary reducer drive shaft (input end) under normal condition

下载: 全尺寸图片幻灯片

图 4 断齿状态下行星减速器传动轴（输入端）幅值

Figure 4. Amplitude diagram of transmission shaft (input end) of planetary reducer under broken tooth condition

下载: 全尺寸图片幻灯片

图 5 正常状态下行星减速器传动轴（输出端）幅值

Figure 5. Amplitude diagram of planetary reducer drive shaft (output) under normal condition

下载: 全尺寸图片幻灯片

图 6 断齿状态下行星减速器传动轴（输出端）幅值

Figure 6. Amplitude diagram of transmission shaft (output end) of planetary reducer under broken tooth condition

下载: 全尺寸图片幻灯片

图 7 截割部某时刻真实运行数据与虚拟运行数据对比

Figure 7. Comparison of real operation data and virtual operation data of cutting part at a certain time

下载: 全尺寸图片幻灯片

图 8 截割部行星减速器传动轴（输入端）加速度虚实数据相似关联融合

Figure 8. Similar correlation fusion of virtual and real acceleration data of transmission shaft (input end ) of planetary reducer of cutting part

下载: 全尺寸图片幻灯片

图 9 截割部行星减速器传动轴（输入端）加速度虚实数据互补关联融合

Figure 9. Complementary correlation fusion of virtual and real acceleration data of transmission shaft (input end ) of planetary reducer of cutting part

下载: 全尺寸图片幻灯片

图 10 基于改进SOM神经网络的故障预警模型

Figure 10. Fault early warning model based on improved self-organizing map neural network

下载: 全尺寸图片幻灯片

图 11 DDS实验平台

Figure 11. Drivetrain diagnostics simulator experimental platform

下载: 全尺寸图片幻灯片

图 12 不同迭代次数下的权值向量

Figure 12. Weight vector graph of different iteration times

下载: 全尺寸图片幻灯片

图 13 训练500次的样本聚类结果

Figure 13. Sample clustering result of 500 training times

下载: 全尺寸图片幻灯片

图 14 SOM神经网络相邻权重之间的距离曲线

Figure 14. Distance curve between adjacent weights of self-organizing map neural network

下载: 全尺寸图片幻灯片

图 15 改进前后不同迭代次数的平均运行时间

Figure 15. Average running time of different iterations before and after improvement

下载: 全尺寸图片幻灯片

表 1 余弦相似度对应变量相关性

Table 1. Cosine similarity corresponding variable correlation

余弦相似度	变量相关性
$0 \leqslant \cos ( {\boldsymbol{X}}, {\boldsymbol{Y}}) ＜ 0.4$	无关
$0.4 \leqslant \cos ( {\boldsymbol{X}}, {\boldsymbol{Y}}) ＜ 0.8$	相关
$0.8 \leqslant \cos ( {\boldsymbol{X}}, {\boldsymbol{Y}}) ＜ 1$	强相关

下载: 导出CSV

表 2 悬臂式掘进机截割部故障类型

Table 2. Fault types of cutting part of cantilever roadheader

截割部部位	故障类型
截割头	截割头转速
行星减速器	传动轴（输入端）加速度
	传动轴（输出端）加速度
	轴承加速度
	太阳轮加速度
截割电动机	截割电动机转速

下载: 导出CSV

表 3 部分正常样本数据

Table 3. Partial sample data

截割头转速/ (m·s⁻¹)	输入轴加速度/ (m·s⁻²)	输出轴加速度/ (m·s⁻²)	轴承加速度/ (m·s⁻²)	太阳轮加速度/ (m·s⁻²)	截割电动机转速/ (m·s⁻²)
0.423	0.003	0.003	−0.073	0.285	1.000
0.442	0	−0.001	−0.151	0.732	0.960
0.426	−0.032	0.001	−0.115	0.978	0.040
0.430	−0.011	0.003	−0.140	0.820	0.560
0.434	0.037	0	−0.167	−0.080	0.600
0.434	0.111	0.004	−0.097	−0.104	0.320
0.438	−0.006	0.003	−0.218	−0.200	0.640
0.446	0.121	0.002	−0.149	−0.292	0.720
0.450	0.092	−0.002	−0.155	−0.728	0.240
0.453	0.057	0.004	−0.176	−0.268	0.040

下载: 导出CSV

表 4 单一故障测试

Table 4. Single fault test

类别	截割头	输入轴	输出轴	轴承	太阳轮	截割电动机
样本获胜神经元	2	10	51	91	100	56
测试获胜神经元	—	10	—	—	—	—

下载: 导出CSV

表 5 复合故障测试

Table 5. Composite fault testing

类别	截割头	输入轴	输出轴	轴承	太阳轮	截割电动机
样本获胜神经元	2	10	51	91	100	56
测试获胜神经元	—	10	—	—	99	—

下载: 导出CSV

表 6 改进前后SOM神经网络不同迭代次数下的运行时间

Table 6. Running time of self-organizing map neural network under different iteration times before and after improvement

迭代次数		单次运行时间/s					平均时间/s
迭代次数		第1次	第2次	第3次	第4次	第5次	平均时间/s
10	改进前	0.0536	0.0618	0.0532	0.0534	0.0540	0.0552
10	改进后	0.0532	0.0434	0.0384	0.0412	0.0454	0.0443
100	改进前	0.2098	0.1963	0.2073	0.1994	0.2149	0.2055
100	改进后	0.1297	0.1236	0.1365	0.1357	0.1257	0.1302
200	改进前	0.3976	0.3642	0.3660	0.3682	0.3700	0.3732
200	改进后	0.2154	0.1904	0.2105	0.2038	0.2066	0.2053
500	改进前	0.8637	0.8571	0.8339	0.8337	0.8385	0.8454
500	改进后	0.4839	0.4990	0.4902	0.4939	0.4837	0.4901

下载: 导出CSV

参考文献(22)

[1]	毛君,董钰峰,卢进南,等. 巷道掘进截割钻进先进技术研究现状及展望[J]. 煤炭学报,2021,46(7):2084-2099. MAO Jun,DONG Yufeng,LU Jinnan,et al. Research status and prospect of advanced technology of roadway excavation cutting and drilling equipment[J]. Journal of China Coal Society,2021,46(7):2084-2099.
[2]	耿晋杰. 掘进机截割部故障分析与优化设计[J]. 西部探矿工程,2020,32(9):160-161. GENG Jinjie. Fault analysis and optimization design of roadheader cutting unit[J]. West-China Exploration Engineering,2020,32(9):160-161.
[3]	吴晋军. 悬臂式掘进机液压故障诊断系统设计[J]. 液压气动与密封,2022,42(8):86-90. WU Jinjun. Design of hydraulic fault diagnosis system for roadheader[J]. Hydraulics Pneumatics & Seals,2022,42(8):86-90.
[4]	盖巍巍,席亚兵,樊志家. 掘进机回转台开裂故障分析及改进[J]. 矿山机械,2021,49(3):18-21. GAI Weiwei,XI Yabing,FAN Zhijia. Analysis on cracking fault of rotary table of roadheader and improvement[J]. Mining & Processing Equipment,2021,49(3):18-21.
[5]	吉昌生. 煤矿井下掘进机机电设备故障诊断及维护[J]. 采矿技术,2018,18(6):122-123. JI Changsheng. Fault diagnosis and maintenance of mechanical and electrical equipment of underground roadheader in coal mine[J]. Mining Technology,2018,18(6):122-123.
[6]	蒙康,滕伟,彭迪康,等. 运行机理与数据双驱动的风电齿轮箱系统故障预警[J]. 中国机械工程,2023,34(12):1476-1485. MENG Kang,TENG Wei,PENG Dikang,et al. Operating mechanism and data driven approach for fault alarm of wind turbine gearbox systems[J]. China Mechanical Engineering,2023,34(12):1476-1485.
[7]	刘送永,刘强. 巷道掘进机健康管理研究现状及展望[J]. 工矿自动化,2021,47(2):32-37. LIU Songyong,LIU Qiang. Research status and prospect of roadheader health management[J]. Industry and Mine Automation,2021,47(2):32-37.
[8]	张天瑞,魏铭琦,刘彬. 全断面掘进机健康管理系统的建模与仿真[J]. 中国工程机械学报,2019,17(3):231-237. ZHANG Tianrui,WEI Mingqi,LIU Bin. System modeling and simulation for health management of tunnel boring machine[J]. Chinese Journal of Construction Machinery,2019,17(3):231-237.
[9]	张境麟,姚钰鹏,冯银辉,等. 故障诊断预警系统在煤炭开采的应用[J]. 煤炭科学技术,2021,49(增刊1):175-182. ZHANG Jinglin,YAO Yupeng,FENG Yinhui,et al. Application of fault diagnosis and early warning system in coal mining[J]. Coal Science and Technology,2021,49(S1):175-182.
[10]	韩宝宏,闫明胜,段鹏飞,等. 基于SOM−MQE模型的设备故障预警方法[J]. 工业技术创新,2021,8(1):74-78. HAN Baohong,YAN Mingsheng,DUAN Pengfei,et al. Equipment fault warning method based on SOM-MQE model[J]. Industrial Technology Innovation,2021,8(1):74-78.
[11]	JI Xiaodong, YANG Yang, QU Yuanyuan, et al. Health diagnosis of roadheader based on reference manifold learning and improved K-means[J/OL]. Shock and Vibration, 2021. https://www.hindawi.com/journals/sv/2021/6311795/.
[12]	FAN Yuhao, QIAN Chenying. Research on fault early warning technology of wind turbine main bearing by stacked auto encoder[J/OL]. Journal of Physics: Conference Series, 2021, 2033（1）. https://iopscience.iop.org/article/10.1088/1742-6596/2033/1/012112/pdf.
[13]	王炳效. 掘进机截割头故障分析及改进方案[J]. 价值工程,2015,34(17):115-116. doi: 10.14018/j.cnki.cn13-1085/n.2015.17.045 WANG Bingxiao. Analysis on the cutting head fault of boring machine and the improvement program[J]. Value Engineering,2015,34(17):115-116. doi: 10.14018/j.cnki.cn13-1085/n.2015.17.045
[14]	王福忠,田晓盈,张丽. 悬臂式掘进机截割机构状态诊断策略研究[J]. 计算机仿真,2015,32(5):390-394. doi: 10.3969/j.issn.1006-9348.2015.05.087 WANG Fuzhong,TIAN Xiaoying,ZHANG Li. State diagnosis strategy for boring machine cutting institutions[J]. Computer Simulation,2015,32(5):390-394. doi: 10.3969/j.issn.1006-9348.2015.05.087
[15]	郝锦龙,董洪全. 掘进机截割部减速器动力学建模及行星架结构强度分析[J]. 煤矿机械,2020,41(10):87-90. doi: 10.13436/j.mkjx.202010027 HAO Jinlong,DONG Hongquan. Dynamic modeling and strength analysis of planet frame structure of roadheader cutting part reducer[J]. Coal Mine Machinery,2020,41(10):87-90. doi: 10.13436/j.mkjx.202010027
[16]	孙英. 掘进机整机动力学分析及电控箱减振研究[D]. 太原: 太原理工大学, 2016. SUN Ying. Research on dynamical analysis of road-header and vibration reduction about the electric cabinet[D]. Taiyuan: Taiyuan University of Technology, 2016.
[17]	贾科,郑黎明,毕天姝,等. 基于余弦相似度的风电场站送出线路纵联保护[J]. 中国电机工程学报,2019,39(21):6263-6275. JIA Ke,ZHENG Liming,BI Tianshu,et al. Pilot protection based on cosine similarity for transmission line connected to wind farms[J]. Proceedings of the CSEE,2019,39(21):6263-6275.
[18]	耿志强,胡海霞,韩永明. 基于余弦相似度的复杂网络故障检测方法及应用[J]. 北京化工大学学报(自然科学版),2017,44(2):87-94. GENG Zhiqiang,HU Haixia,HAN Yongming. A complex network fault detection method based on cosine similarity and its application[J]. Journal of Beijing University of Chemical Technology(Natural Science Edition),2017,44(2):87-94.
[19]	胡津铭,赵戈,张艳. 基于改进SOM神经网络的移动警务终端流量监测方法研究[J]. 信息网络安全,2021(增刊1):109-112. HU Jinming,ZHAO Ge,ZHANG Yan. Research on the traffic monitoring method of mobile police terminal based on improved SOM neural network[J]. Netinfo Security,2021(S1):109-112.
[20]	申屠晗,彭冬亮,薛安克. 最小熵反馈式变结构多模型融合算法[J]. 控制理论与应用,2013,30(3):372-378. SHEN Tuhan,PENG Dongliang,XUE Anke. Minimum entropy and feedback structure-based algorithm for variable structure multi-model fusion[J]. Control Theory & Applications,2013,30(3):372-378.
[21]	刘达,陈松灿. 基于高斯神经元的自组织映射网络研究[J]. 数据采集与处理,2023,38(1):85-92. LIU Da,CHEN Songcan. Research on self-organizing map network based on gaussian neuron[J]. Journal of Data Acquisition and Processing,2023,38(1):85-92.
[22]	吴钱昊. 基于数字孪生车间的生产车间设备监控和设备故障预警方法研究[D]. 合肥: 合肥工业大学, 2020. WU Qianhao. Research on methods for equipment monitoring and equipment fault early warning of production workshop based on digital twin workshop[D]. Hefei: Hefei University of Technology, 2020.