Automatic vibration fault detection of coal mine explosion-proof electrical equipment based on One-Class Support Vector Machine
-
摘要:
煤矿防爆电气设备在运行过程中产生的振动会损害其机械完整性，导致紧固件松动、零部件磨损，并改变设备的结构与振动模态，进而引发信号特征的复杂变化，使得正常振动频率与故障引发的新频率成分相互混淆，模糊了正常信号与故障信号之间的界限，从而降低传统检测方法在故障检测中的准确性。针对上述问题，提出一种基于单分类支持向量机（OCSVM）的煤矿防爆电气设备振动故障自动检测方法。首先，构造设备的正常状态特征和振动故障状态特征，根据OCSVM的特性，将正常状态特征序列设定为OCSVM核函数的决策边界学习目标。考虑煤矿防爆电气设备振动故障信号呈现非线性和高维特征，选定多项式核作为OCSVM的核函数。然后，采用网格搜索和K−交叉验证相结合的方式对OCSVM进行参数调优，以使OCSVM达到更好的性能。最后，通过求取OCSVM目标函数的最优解，确定最优决策边界，以此实现煤矿防爆电气设备振动故障的自动检测。实验结果显示：① 在迭代次数为20时，OCSVM算法算法可完成收敛，达到稳定。② 基于OCSVM的电气设备信号划分实验中，借助多项式核函数能精准划分样本实现检测。③ 振动故障自动检测性能分析中，所提方法在各样本量下准确率均显著高于红外热成像技术检测方法、基于灰狼优化支持向量机模型检测方法，小样本量时准确率达98.25%且稳定性好。
Abstract:The vibration generated by explosion-proof electrical equipment in coal mines during operation can compromise its mechanical integrity, leading to fastener loosening, component wear, and changes in the structure and vibration modes of the equipment. This can cause complex changes in signal features, resulting in confusion between normal vibration frequency and new frequency components induced by faults. As a result, the boundary between normal and fault signals becomes unclear, reducing the accuracy of traditional fault detection methods. To address this issue, an automatic vibration fault detection method for coal mine explosion-proof electrical equipment was proposed based on One-Class Support Vector Machine (OCSVM). First, the normal state features and vibration fault state features of the equipment were constructed. Based on the characteristics of OCSVM, the normal state feature sequence was set as the learning target for the decision boundary of the OCSVM kernel function. Due to the nonlinear and high-dimensional characteristics of vibration faults in explosion-proof electrical equipment, a polynomial kernel was selected as the OCSVM kernel function after comprehensive consideration. Then, grid search combined with K-fold cross-validation was used to optimize the parameters of the OCSVM, ensuring better performance. Finally, by obtaining the optimal solution of the OCSVM objective function, the optimal decision boundary was determined to realize automatic fault detection of vibration faults in coal mine explosion-proof electrical equipment. Experimental results showed that: ① When the number of iterations is 20, the OCSVM algorithm can complete convergence and achieve stability. ② In the electrical equipment signal classification experiment based on OCSVM, the use of the polynomial kernel function accurately classified samples for detection. ③ In the performance analysis of automatic vibration fault detection, the proposed method showed significantly higher accuracy across different sample sizes than infrared thermography and detection methods based on grey wolf optimization and support vector machine. Under small sample sizes, it achieved an accuracy of 98.25% with good stability.
-
-
![]()
图 2 基于OCSVM的电气设备信号划分结果
Figure 2. Signal classification results of electrical equipment based on OCSVM
表 1 矿用防爆电动机技术参数
Table 1 Technical specification of mine explosion-proof motor
参数 值 参数 值 额定电压/V 380 同步转速/（r·min−1） 1 440 额定电流/A 11.37 额定转速/（r·min−1） 1 460 额定频率/Hz 50 效率/% 89.6 额定功率/kW 5.5 功率因数 0.82 额定转矩/(N·m) 36.48 电源频率/Hz 50 
下载: 导出CSV
-
[1] 雷志鹏,姜宛廷,门汝佳,等. 矿用三元乙丙橡胶高压电缆绝缘老化机理及状态评估技术研究进展[J]. 工矿自动化,2023,49(9):167-177. LEI Zhipeng,JIANG Wanting,MEN Rujia,et al. Research progress on insulation aging mechanism and condition evaluation technology of mining EPDM high-voltage cables[J]. Journal of Mine Automation,2023,49(9):167-177.
[2] 刘思莉,张宇,刘军. 地下储气库生产站场防爆电气设备使用与管理[J]. 石油化工安全环保技术,2022,38(6):42-44,7. LIU Sili,ZHANG Yu,LIU Jun. Use and management of explo-sion-proof electrical equipment in production sites of underground gas storage[J]. Petrochemical Safety and Environmental Protection Technology,2022,38(6):42-44,7.
[3] 罗振敏,王晓悦,丁旭涵,等. 碳氢−生物表面活性剂在电解质加载下的降尘及防爆性能研究[J]. 中国安全生产科学技术,2023,19(增刊2):166-173. LUO Zhenmin,WANG Xiaoyue,DING Xuhan,et al. Study on dust reduction and explosion-proof performance of hydrocarbon-bio surfactants under electrolyte loading[J]. Journal of Safety Science and Technology,2023,19(S2):166-173.
[4] LAIB DIT LEKSIR Y,GUERFI K,AMOURI A,et al. Detection of electrical fault in medium voltage installation using support vector machine and artificial neural network[J]. Russian Journal of Nondestructive Testing,2022,58(3):176-185. DOI: 10.1134/S1061830922030081
[5] ZOU Xin,LYU Rongxin,LI Xinyan,et al. Intelligent electrical fault detection and recognition based on gray wolf optimization and support vector machine[J]. Journal of Physics:Conference Series,2022,2181(1). DOI: 10.1088/1742-6596/2181/1/012058.
[6] SHI Wenyun,REN Xiaoming. Electrical fault detection equipment based on infrared image fusion[J]. Procedia Computer Science,2022,208:509-515. DOI: 10.1016/j.procs.2022.10.070
[7] CHELLAMUTHU S,SEKARAN E C,ANNAMALAI S,et al. Fault detection in electrical equipment by infrared thermography images using spiking neural network through hybrid feature selection[J]. Journal of Circuits, Systems and Computers,2022, 32(8) . DOI: 10.1142/S0218126623501396.
[8] 刘赫,赵天成,刘俊博,等. 基于深度残差UNet网络的电气设备红外图像分割方法[J]. 红外技术,2022,44(12):1351-1357. LIU He,ZHAO Tiancheng,LIU Junbo,et al. Deep residual UNet network-based infrared image segmentation method for electrical equipment[J]. Infrared Technology,2022,44(12):1351-1357.
[9] 邓军,王志强,王伟峰,等. 基于LSTM−AE−OCSVM的带式输送机火灾监测隐患识别技术[J]. 煤炭技术,2023,42(1):225-229. DENG Jun,WANG Zhiqiang,WANG Weifeng,et al. Hidden danger identification technology of belt conveyor fire monitoring based on LSTM-AE-OCSVM[J]. Coal Technology,2023,42(1):225-229.
[10] 黄宇斐,石新发,贺石中,等. 一种基于主成分分析与支持向量机的风电齿轮箱故障诊断方法[J]. 热能动力工程,2022,37(10):175-181. HUANG Yufei,SHI Xinfa,HE Shizhong,et al. A fault diagnosis method of wind turbine gearbox based on PCA and SVM[J]. Journal of Engineering for Thermal Energy and Power,2022,37(10):175-181.
[11] 闫汇聪,刘德山,陈浪,等. 散度核协同表示与空谱融合特征的高光谱图像分类算法[J]. 计算机应用与软件,2023,40(2):287-295. DOI: 10.3969/j.issn.1000-386x.2023.02.045 YAN Huicong,LIU Deshan,CHEN Lang,et al. A hyperspectral image classification algorithm based on divergence kernel collaborative representation and spatial-spectral fusion feature[J]. Computer Applications and Software,2023,40(2):287-295. DOI: 10.3969/j.issn.1000-386x.2023.02.045
[12] 余柏杨,吕宏强,周岩,等. 基于机器学习的高速复杂流场流动控制效果预测分析[J]. 实验流体力学,2022,36(3):44-54. DOI: 10.11729/syltlx20210168 YU Baiyang,LYU Hongqiang,ZHOU Yan,et al. Predictive analysis of flow control in high-speed complex flow field based on machine learning[J]. Journal of Experiments in Fluid Mechanics,2022,36(3):44-54. DOI: 10.11729/syltlx20210168
[13] 金长宇,于佳强,王强,等. 基于集成学习CatBoost优化模型的爆堆大块率预测[J]. 东北大学学报(自然科学版),2023,44(12):1743-1750. JIN Changyu,YU Jiaqiang,WANG Qiang,et al. Prediction of blasting fragment large block percentage ratio based on ensemble learning CatBoost model[J]. Journal of Northeastern University(Natural Science),2023,44(12):1743-1750.
[14] 花靖,蒋秀,于超,等. 基于改进型SVM算法的气液两相流持液率计算模型[J]. 西安石油大学学报(自然科学版),2022,37(6):103-110,118. DOI: 10.3969/j.issn.1673-064X.2022.06.014 HUA Jing,JIANG Xiu,YU Chao,et al. Liquid holdup calculation model of gas-liquid two-phase flow based on improved SVM algorithm[J]. Journal of Xi'an Shiyou University(Natural Science Edition),2022,37(6):103-110,118. DOI: 10.3969/j.issn.1673-064X.2022.06.014
[15] 荣统瑞,侯恩科,夏冰冰. 基于二次分解和BO-BiLSTM组合模型的采煤工作面瓦斯涌出量预测方法研究[J]. 煤矿安全,2024,55(5):83-92. RONG Tongrui,HOU Enke,XIA Bingbing. Research on prediction method of coal mining face gas outflow based on quadratic decomposition and BO-BiLSTM combination model[J]. Safety in Coal Mines,2024,55(5):83-92.
[16] 师素珍,石贵飞,刘最亮,等. 基于多变量LSTM网络的K2灰岩富水区预测——以阳泉泊里矿区为例[J]. 煤田地质与勘探,2023,51(5):155-163. SHI Suzhen,SHI Guifei,LIU Zuiliang,et al. Predicting the water-yield properties of K2 limestones based on multivariate LSTM neural network:a case study of the Poli Mining Area in Yangquan[J]. Coal Geology & Exploration,2023,51(5):155-163.
[17] 代鑫,胡斌,李京,等. 炭质泥页岩剪切破坏声发射特性及其分形特征[J]. 科学技术与工程,2024,24(12):4909-4915. DAI Xin,HU Bin,LI Jing,et al. Acoustic emission of carbonaceous shale and its fractal characteristics under shear failure[J]. Science Technology and Engineering,2024,24(12):4909-4915.
[18] XU Weihua,BU Qinyuan. Matrix-based incremental feature selection method using weight-partitioned multigranulation rough set[J]. Information Sciences,2024,681. DOI: 10.1016/j.ins.2024.121219.
[19] 王静红,田长申,李昊康,等. 基于拉格朗日对偶的小样本学习隐私保护和公平性约束方法[J]. 计算机科学,2024,51(7):405-412. DOI: 10.11896/jsjkx.230500012 WANG Jinghong,TIAN Changshen,LI Haokang,et al. Lagrangian dual-based privacy protection and fairness constrained method for few-shot learning[J]. Computer Science,2024,51(7):405-412. DOI: 10.11896/jsjkx.230500012
[20] 张先锋. 具终端状态约束的无穷维随机发展方程的线性二次最优控制[J]. 四川大学学报(自然科学版),2024,61(3):75-80. ZHANG Xianfeng. Linear quadratic optimal control problem for stochastic evolution equations with terminal state constraints in infinite dimensions[J]. Journal of Sichuan University(Natural Science Edition),2024,61(3):75-80.
[21] 林韧昊,周清雷,扈天卿,等. 基于决策边界分析的深度神经网络鲁棒性评估与优先次序验证[J]. 计算机学报,2024,47(4):862-876. LIN Renhao,ZHOU Qinglei,HU Tianqing,et al. Robustness evaluation and prioritization verification for deep neural networks via decision boundary analysis[J]. Chinese Journal of Computers,2024,47(4):862-876.
-
期刊类型引用(16)
1. 宋庆军，焦守悦，姜海燕，宋庆辉，郝文超. 基于改进EfficientNet的煤矸音频分类方法. 工矿自动化. 2025(01): 138-144 . 
本站查看
2. 燕建华. 周期性冲击波形匹配下选煤破碎机滚动轴承局部缺陷检测. 自动化与仪器仪表. 2025(02): 37-41 . 百度学术
3. 上官星驰，张晓良，刘朝，石会，王嘉宇. 基于改进特征提取算法及胶囊网络的设备故障诊断研究. 工矿自动化. 2024(S1): 146-150 . 本站查看
4. 范忠明. 基于神经网络图像识别技术的放顶煤煤矸自动识别方法. 自动化技术与应用. 2024(10): 39-42 . 百度学术
5. 王志峰，王家臣，李良晖，安博超. 基于DeepLab v3+的综放工作面含矸率预测研究. 工矿自动化. 2024(10): 90-96 . 本站查看
6. 李立宝，袁永，秦正寒，李波，闫政天，李勇. 图像特征与振动频谱多源融合驱动的煤矸识别技术研究. 工矿自动化. 2024(11): 43-51 . 本站查看
7. 周正南，刘美，吴斌鑫，莫常春，高兴泉，张斐. 基于改进的CEEMDAN与关联维数的石化轴承故障特征提取. 机床与液压. 2023(05): 212-217 . 百度学术
8. 司垒，李嘉豪，邢峰，魏东，戴剑博，王忠宾. 不同煤矸混合物的微波传播特性试验研究. 煤炭科学技术. 2023(05): 219-231 . 百度学术
9. 石港，雷志鹏. 基于改进深度森林的采煤机拖拽电缆挤压力识别方法. 工矿自动化. 2023(10): 8-16+51 . 本站查看
10. 史翔予，司垒，王忠宾，魏东，顾进恒. 基于改进双向峰-谷搜索算法的煤矸模型电磁波正演模拟. 工矿自动化. 2023(10): 87-95 . 本站查看
11. 李春锋，马星河，刘广朋. 基于改进VMD的矿用电缆局放信号降噪方法. 能源与环保. 2023(12): 268-274 . 百度学术
12. 贺海涛，王佳豪，张海峰，荣耀，崔耀. 基于U-Net的放煤状态控制关键技术研究. 煤炭科学技术. 2022(S2): 237-243 . 百度学术
13. 高丰，朱少成，罗石. 基于改进的经验模态分解的后视镜驱动器故障诊断方法. 河南科技大学学报(自然科学版). 2021(06): 39-45+6-7 . 百度学术
14. 刘丹丹. 基于EMD的GNSS时间序列异常值探测算法. 地球物理学进展. 2021(05): 1865-1873 . 百度学术
15. 丁震，常博深. 面向煤矸识别的近红外反射光谱数据预处理方法. 工矿自动化. 2021(12): 93-97 . 本站查看
16. 田志飞，洪盛勇. 火力发电厂煤流状态实时监测方法的研究及应用. 电力学报. 2021(06): 564-572 . 百度学术
其他类型引用(14)
计量
- 文章访问数: 30
- HTML全文浏览量: 3
- PDF下载量: 6
- 被引次数: 30
