基于参数优化的VMD-SVD和LSTM的输电杆塔倾斜状态识别

doi:10.11930/j.issn.1004-9649.202302065

摘要/Abstract

摘要：

针对输电杆塔结构状态信息提取难度大、精度低等问题，提出了一种基于北方苍鹰算法优化的变分模态分解（northern goshawk optimized variational mode decomposition，NGO-VMD）与长短期记忆（long short-term memory，LSTM）神经网络的输电杆塔倾斜状态识别方案。通过北方苍鹰优化算法解决了变分模态分解参数难确定的问题，并且证明其分解的各阶本征模态分量（intrinsic mode function，IMF）可以有效提取出杆塔结构的模态信息。为了使信息特征更为明显，对IMF分量进行奇异值分解（singular value decomposition，SVD），发现各阶分量的奇异值在杆塔不同状态下有较为明显的区别。最后引入LSTM神经网络进行特征分类，形成故障诊断模型。依托某110 kV猫头塔对模型进行试验验证，结果表明：所提方法对杆塔倾斜状态的识别准确率为96.68%，与其他方法相比，具有效率更高、稳定性更强、更加精准的优势。

关键词: 杆塔倾斜, 状态识别, 北方苍鹰算法优化, 自适应变分模态分解, 奇异值分解, 长短期记忆神经网络

Abstract:

To address the problems of high difficulty and poor accuracy in extracting the structural state information of transmission towers, a transmission tower tilt state recognition solution is proposed based on the northern goshawk optimized variational mode decomposition (NGO-VMD) and long short-term memory (LSTM) neural network. The problem to determine the VMD parameters is solved by NGO, and it is demonstrated that the decomposed intrinsic mode function (IMF) components of each order can effectively extract the modal information of the tower structure. In order to make the information features more obvious, the singular value decomposition (SVD) of IMF components is performed, and it is found that the singular values of each order component have more obvious differences in different states of the tower. Finally, the LSTM neural network is introduced for feature classification to form a fault diagnosis model. A 110 kV cathead-type tower is used to verify the proposed model, and the results show that the proposed method can achieve an accuracy of 96.68% in identification of tower tilting state. Compared with other methods, this solution has the advantages of higher efficiency, stronger stability and better accuracy.

Key words: tower tilt, state recognition, northern goshawk optimized, adaptive variational mode decomposition, singular value decomposition, long short-term memory neural network

赵隆, 温冠儒, 刘志成, 袁鹏, 董新胜. 基于参数优化的VMD-SVD和LSTM的输电杆塔倾斜状态识别[J]. 中国电力, 2023, 56(12): 217-226, 237.

Long ZHAO, Guanru WEN, Zhicheng LIU, Peng YUAN, Xinsheng DONG. Transmission Tower Tilt State Recognition Based on Parameter Optimization of VMD-SVD and LSTM[J]. Electric Power, 2023, 56(12): 217-226, 237.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202302065

https://www.electricpower.com.cn/CN/Y2023/V56/I12/217

图/表 17

图 1 杆塔倾斜状态识别流程

Fig.1 Tower tilt state recognition process

图 2 NGO-VMD流程

Fig.2 NGO-VMD Process

图 3 LSTM单元结构

Fig.3 LSTM cell structure

图 4 有限元模拟

Fig.4 Finite element simulation

表 1 杆塔固有频率

Table 1 Tower inherent frequency

模态数		固有频率/Hz
1		2.5770
2		3.2770
3		9.2728
4		14.8530
5		17.2590

图 5 NGO-VMD与CEEMDAN的分解结果

Fig.5 Decomposition results of NGO-VMD and CEEMDAN

图 6 NGO-VMD与CEEMDAN分解的IMF频域分量

Fig.6 IMF frequency domain components of NGO-VMD and CEEMDAN decomposition

图 7 输电杆塔第3阶振型

Fig.7 The third-order vibration mode of transmission tower

图 8 响应测试现场

Fig.8 Response test site

图 9 猫头塔振动信号

Fig.9 Cathead tower vibration signals

图 10 各优化算法适应度曲线

Fig.10 Adaptation curve of each optimization algorithm

图 11 振动信号的NGO-VMD分解

Fig.11 NGO-VMD decomposition of vibration signals

图 12 不同倾斜状态下的奇异值分布

Fig.12 Distribution of singular values in different tilt states

图 13 第3、4阶奇异值箱图

Fig.13 The 3rd and 4th order singular value box plots

图 14 LSTM训练进度

Fig.14 LSTM training progress

图 15 不同分类方法的测试集分类结果

Fig.15 Classification results of test sets with different classification methods

表 2 不同分类方法的准确率

Table 2 Accuracy of different classification methods

分类方法	训练集准确率/%	测试集准确率/%
LSTM	98.89	96.67
BP	95.56	88.34
RF	94.45	91.67
SVM	84.45	76.67

参考文献 25

1	DONG X S, WEN G R, ZHAO M G, et al. Study of the prevention method of ±800 kV transmission tower foundation deviation[J]. Energies, 2023, 16 (6): 2557. DOI
2	黄新波, 廖明进, 徐冠华, 等. 采用光纤光栅传感器的输电线路铁塔应力监测方法[J]. 电力自动化设备, 2016, 36 (4): 68- 72.
	HUANG Xinbo, LIAO Mingjin, XU Guanhua, et al. Stress monitoring method applying FBG sensor for transmission line towers[J]. Electric Power Automation Equipment, 2016, 36 (4): 68- 72.
3	LU Z M, GONG H, JIN Q H, et al. A transmission tower tilt state assessment approach based on dense point cloud from UAV-based LiDAR[J]. Remote Sensing, 2022, 14 (2): 408. DOI
4	高雪莲, 李乐依, 刘文琳. 基于遥感卫星光学影像的杆塔倾斜在线监测方法[J]. 中国电机工程学报, 2022, 42 (16): 5971- 5978, 6171.
	GAO Xuelian, LI Leyi, LIU Wenlin. On-line monitoring method of pole and tower tilt based on remote sensing satellite optical image[J]. Proceedings of the CSEE, 2022, 42 (16): 5971- 5978, 6171.
5	WU B L, TONG L, CHEN Y. Revised improved DINSAR algorithm for monitoring the inclination displacement of top position of electric power transmission tower[J]. IEEE Geoscience and Remote Sensing Letters, 2018, 15 (6): 877- 881. DOI
6	何冰, 王媚, 张羽公, 等. 基于振动信号的输电铁塔螺栓松动检测技术研究[J]. 功能材料与器件学报, 2019, 25 (3): 214- 219.
	HE Bing, WANG Mei, ZHANG Yugong, et al. Research on loose detection technology of transmission tower bolt based on vibration signal[J]. Journal of Functional Materials and Devices, 2019, 25 (3): 214- 219.
7	张安安, 邓芳明. 基于CNN-RVM的输电杆塔外破振动识别方案[J]. 计算机仿真, 2020, 37 (4): 76- 80.
	ZHANG Anan, DENG Fangming. External vibration identification scheme of transmission line tower based on CNN-RVM[J]. Computer Simulation, 2020, 37 (4): 76- 80.
8	DRAGOMIRETSKIY K, ZOSSO D. Variational mode decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62 (3): 531- 544. DOI
9	贾亚飞, 朱永利, 王刘旺, 等. 基于VMD和多尺度熵的变压器内绝缘局部放电信号特征提取及分类[J]. 电工技术学报, 2016, 31 (19): 208- 217.
	JIA Yafei, ZHU Yongli, WANG Liuwang, et al. Feature extraction and classification on partial discharge signals of power transformers based on VMD and multiscale entropy[J]. Transactions of China Electrotechnical Society, 2016, 31 (19): 208- 217.
10	唐贵基, 王晓龙. 参数优化变分模态分解方法在滚动轴承早期故障诊断中的应用[J]. 西安交通大学学报, 2015, 49 (5): 73- 81.
	TANG Guiji, WANG Xiaolong. Parameter optimized variational mode decomposition method with application to incipient fault diagnosis of rolling bearing[J]. Journal of Xi’an Jiaotong University, 2015, 49 (5): 73- 81.
11	冯玉芳, 卢厚清, 殷宏, 等. 基于BP神经网络的故障诊断模型研究[J]. 计算机工程与应用, 2019, 55 (6): 24- 30. DOI
	FENG Yufang, LU Houqing, YIN Hong, et al. Study on fault diagnosis model based on BP neural network[J]. Computer Engineering and Applications, 2019, 55 (6): 24- 30. DOI
12	BAI S, TANG H D, AN S. Coordinate CNNs and LSTMs to categorize scene images with multi-views and multi-levels of abstraction[J]. Expert Systems With Applications, 2019, 120, 298- 309. DOI
13	王立林, 刘俊. 基于多尺度卷积的船舶行为识别方法[J]. 计算机应用, 2019, 39 (12): 3691- 3696.
	WANG Lilin, LIU Jun. Ship behavior recognition method based on multi-scale convolution[J]. Journal of Computer Applications, 2019, 39 (12): 3691- 3696.
14	肖雄, 王健翔, 张勇军, 等. 一种用于轴承故障诊断的二维卷积神经网络优化方法[J]. 中国电机工程学报, 2019, 39 (15): 4558- 4568.
	XIAO Xiong, WANG Jianxiang, ZHANG Yongjun, et al. A two-dimensional convolutional neural network optimization method for bearing fault diagnosis[J]. Proceedings of the CSEE, 2019, 39 (15): 4558- 4568.
15	RAVIKUMAR K N, YADAV A, KUMAR H, et al. Gearbox fault diagnosis based on Multi-Scale deep residual learning and stacked LSTM model[J]. Measurement, 2021, 186, 110099. DOI
16	DEHGHANI M, HUBÁLOVSKÝ Š, TROJOVSKÝ P. Northern goshawk optimization: a new swarm-based algorithm for solving optimization problems[J]. IEEE Access, 2021, 9, 162059- 162080. DOI
17	YANG W X, TSE P W. Development of an advanced noise reduction method for vibration analysis based on singular value decomposition[J]. NDT & E International, 2003, 36 (6): 419- 432.
18	于洋, 何明, 刘博, 等. 基于TL-LSTM的轴承故障声发射信号识别研究[J]. 仪器仪表学报, 2019, 40 (5): 51- 59.
	YU Yang, HE Ming, LIU Bo, et al. Research on acoustic emission signal recognition of bearing fault based on TL-LSTM[J]. Chinese Journal of Scientific Instrument, 2019, 40 (5): 51- 59.
19	BIE F F, DU T F, LYU F X, et al. An integrated approach based on improved CEEMDAN and LSTM deep learning neural network for fault diagnosis of reciprocating pump[J]. IEEE Access, 2021, 9, 23301- 23310. DOI
20	郑天堂, 黄新波, 赵隆, 等. 输电杆塔沉降状态感知技术研究[J]. 广东电力, 2021, 34 (4): 85- 93. DOI
	ZHENG Tiantang, HUANG Xinbo, ZHAO Long, et al. Research on sensing technology of transmission tower settlement state[J]. Guangdong Electric Power, 2021, 34 (4): 85- 93. DOI
21	KARAMI-MOHAMMADI R, MIRTAHERI M, SALKHORDEH M, et al. Vibration anatomy and damage detection in power transmission towers with limited sensors[J]. Sensors, 2020, 20 (6): 1731. DOI
22	LEE C Y, HUANG N E, CHERN C C, et al. A study on the dynamic characteristics and ultimate capacity of steel transmission towers[J]. Journal of the Chinese Institute of Engineers, 2013, 36 (3): 273- 294. DOI
23	ZHANG Q, FU X, REN L, et al. Modal parameters of a transmission tower considering the coupling effects between the tower and lines[J]. Engineering Structures, 2020, 220, 110947. DOI
24	GAI J B, SHEN J X, HU Y F, et al. An integrated method based on hybrid grey wolf optimizer improved variational mode decomposition and deep neural network for fault diagnosis of rolling bearing[J]. Measurement, 2020, 162, 107901. DOI
25	LIU H M, WANG X A, LU C. Rolling bearing fault diagnosis under variable conditions using Hilbert-Huang transform and singular value decomposition[J]. Mathematical Problems in Engineering, 2014, 2014, 1- 10.

[1]	邵必林, 纪丹阳. 基于VMD-SE的电力负荷分量的多特征短期预测[J]. 中国电力, 2024, 57(4): 162-170.
[2]	陈庆斌, 杨耿煌, 耿丽清, 苏娟, 孙京生. 基于相似日选取和数据重构的短期光伏功率组合预测方法[J]. 中国电力, 2024, 57(12): 71-81.
[3]	杨晓峰, 方逸航, 赵鹏臻, 王承民, 谢宁. 基于K-means和BPNN的风机状态识别[J]. 中国电力, 2023, 56(6): 158-166,175.
[4]	于虹, 李昊, 康彪, 汪大海, 张志强, 张贵峰, 张海鹏. 采空塌陷区输电杆塔基础变形率限值研究[J]. 中国电力, 2022, 55(12): 124-134.
[5]	黄冬梅, 王玥琦, 胡安铎, 孙锦中, 时帅, 孙园, 房岭锋. 融合多维度特征的绝缘子状态边缘识别方法[J]. 中国电力, 2022, 55(1): 133-141.
[6]	胡翀, 徐斌, 甄超, 赵新德, 王昕, 唐兴勇. 基于电压暂降监测数据的敏感负荷非侵入式识别方法[J]. 中国电力, 2021, 54(8): 35-42,51.