一种基于EGK-均值聚类和PCA的海流发电机叶片冲击故障检测方法

谢涛; 张卫东; 张义博

doi:10.19912/j.0254-0096.tynxb.2022-1793

PDF(5537 KB)

太阳能学报 ›› 2024, Vol. 45 ›› Issue (3) : 201-209. DOI: 10.19912/j.0254-0096.tynxb.2022-1793

一种基于EGK-均值聚类和PCA的海流发电机叶片冲击故障检测方法

谢涛, 张卫东, 张义博

作者信息 +

AN IMPACT FAULT DETECTION METHOD OF MARINE CURRENT TURBINE BLADE VIA EGK-MEANS AND PCA

Xie Tao, Zhang Weidong, Zhang Yibo

Author information +

文章历史 +

摘要

海流发电机受海流流速变化、随机湍流的影响呈现多变工况,想要准确检测出叶片冲击故障是一项巨大挑战。基于此,提出一种基于包络几何特征聚类（EGK-均值聚类）的方法用于划分变工况下产生的定子电流信号,在不同子工况下建立故障检测模型。首先,根据样本矩阵构建包络几何特征矩阵,利用几何特征矩阵确定初始聚类中心进行聚类;然后,基于各工况下的包络数据分别进行主元分析以降低数据维度;最后,T²和SPE统计量被计算用于实时冲击故障检测。一台海流发电机样机及配套循环水槽被搭建验证本文所提方法。实验结果表明本文所提方法在变工况下针对海流发电机叶片冲击故障有着显著的识别能力和检测精度。

Abstract

The operating conditions of MCTs are affected by varying water velocity and random turbulence. The blades of the MCT s are prone to cracks due to long-term exposure to sea water and are quickly impacted by fish or seabed creatures. Frequent changes in marine currents lead to variable working conditions. A detection method based on envelop geometrical K-means （EGK-means） to divide the stator current signals generated under variable conditions and establish fault detection models is proposed. First, construct the envelope geometric feature based on the sample matrix, use the geometric feature matrix to select the initial clustering center for clustering, perform PCA modelling based on the current data under each working condition to reduce the data dimension. Finally, the T² and SPE statistics are calculated for real-time fault detection. A prototype MCT and supporting circulating water tank were built to verify the proposed method. The diagnostic results verify that the proposed method has significant recognition ability and detection accuracy for the impact faults of MCTs under variable marine conditions.

导出引用

谢涛, 张卫东, 张义博. 一种基于EGK-均值聚类和PCA的海流发电机叶片冲击故障检测方法[J]. 太阳能学报. 2024, 45(3): 201-209 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1793

Xie Tao, Zhang Weidong, Zhang Yibo. AN IMPACT FAULT DETECTION METHOD OF MARINE CURRENT TURBINE BLADE VIA EGK-MEANS AND PCA[J]. Acta Energiae Solaris Sinica. 2024, 45(3): 201-209 https://doi.org/10.19912/j.0254-0096.tynxb.2022-1793

中图分类号： TM313

参考文献

[1] 刘玉新, 王海峰, 王冀, 等. 海洋强国建设背景下加快海洋能开发利用的思考[J]. 科技导报, 2018, 36(14): 22-25.
LIU Y X, WANG H F, WANG J, et al.Reflections on the exploitation and utilization of marine renewable energy under the background of ocean power[J]. Science & technology review, 2018, 36(14): 22-25.
[2] 刘星. 海洋能发电前景可期[J]. 电气技术, 2017(5): 5.
LIU X.The prospect of ocean energy power generation can be expected[J]. Electrical engineering, 2017(5): 5.
[3] MARTINEZ R, PAYNE G S, BRUCE T.The effects of oblique waves and currents on the loadings and performance of tidal turbines[J]. Ocean engineering, 2018, 164: 55-64.
[4] 马洁, 胡仁松, 白连平. 潮流发电最大功率跟踪发电机失速问题研究[J]. 电气技术, 2019, 20(11): 49-52, 76.
MA J, HU R S, BAI L P.Research on maximum power tracking generator stall of tidal current generation[J]. Electrical engineering, 2019, 20(11): 49-52, 76.
[5] SONG K, WANG W Q, YAN Y.Experimental and numerical analysis of a multilayer composite ocean current turbine blade[J]. Ocean engineering, 2020, 198: 106977.
[6] 党志高, 毛昭勇, 田文龙. 水下系留平台的海流发电叶片翼型流噪声特性研究[J]. 太阳能学报, 2021, 42(4): 81-88.
DANG Z G, MAO Z Y, TIAN W L.Investigation on noise characteristics of hydrofoils of marine current turbines for underwater mooring platforms[J]. Acta energiae solaris sinica, 2021, 42(4): 81-88.
[7] SONG K, WANG W Q, YAN Y.The hydrodynamic performance of a tidal-stream turbine in shear flow[J]. Ocean engineering, 2020, 199: 107035.
[8] FITURI A, ALY H H, EL-HAWARY M E. Unsteady surface wave influence on tidal current power forecasting[J]. Ocean engineering, 2020, 218: 108231.
[9] XIE T, WANG T Z, HE Q Q, et al.A review of current issues of marine current turbine blade fault detection[J]. Ocean engineering, 2020, 218: 108194.
[10] 徐青山, 娄藕蝶, 郑爱霞, 等. 基于近邻传播聚类和遗传优化的非侵入式负荷分解方法[J]. 电工技术学报, 2018, 33(16): 3868-3878.
XU Q S, LOU O D, ZHENG A X, et al.A non-intrusive load decomposition method based on affinity propagation and genetic algorithm optimization[J]. Transactions of China Electrotechnical Society, 2018, 33(16): 3868-3878.
[11] MAQSOOD A, OSLEBO D, CORZINE K, et al.STFT cluster analysis for DC pulsed load monitoring and fault detection on naval shipboard power systems[J]. IEEE transactions on transportation electrification, 2020, 6(2): 821-831.
[12] 李春兰, 罗杰, 王长云, 等. 基于循环谱特征和聚类分析的触电识别[J]. 电工技术学报, 2021, 36(22): 4677-4687.
LI C L, LUO J, WANG C Y, et al.Electric shock recognition method based on cyclic spectrum features and cluster analysis[J]. Transactions of China Electrotechnical Society, 2021, 36(22): 4677-4687.
[13] 冯立伟, 张成, 李元, 等. 基于权重k近邻的多模态过程故障检测方法[J]. 控制工程, 2019, 26(11): 1986-1993.
FENG L W, ZHANG C, LI Y, et al.Fault detection method based on weighted k nearest neighbor in multimode process[J]. Control engineering of China, 2019, 26(11): 1986-1993.
[14] 江志农, 赵南洋, 夏敏, 等. 一种基于流形学习和KNN算法的柴油机工况识别方法[J]. 噪声与振动控制, 2019, 39(3): 1-6.
JIANG Z N, ZHAO N Y, XIA M, et al.Operating mode identification method for diesel engines based on manifold learning and KNN algorithm[J]. Noise and vibration control, 2019, 39(3): 1-6.
[15] YOO Y.Data-driven fault detection process using correlation based clustering[J]. Computers in industry, 2020, 122: 103279.
[16] 朱永利, 蒋伟, 刘刚. 基于密度峰值聚类算法的局部放电脉冲分割[J]. 电工技术学报, 2020, 35(6): 1377-1386.
ZHU Y L, JIANG W, LIU G.Partial discharge pulse segmentation based on clustering by fast search and find of density peaks[J]. Transactions of China Electrotechnical Society, 2020, 35(6): 1377-1386.
[17] 周东国, 张恒, 周洪, 等. 基于状态特征聚类的非侵入式负荷事件检测方法[J]. 电工技术学报, 2020, 35(21): 4565-4575.
ZHOU D G, ZHANG H, ZHOU H, et al.Non-intrusive load event detection method based on state feature clustering[J]. Transactions of China Electrotechnical Society, 2020, 35(21): 4565-4575.
[18] BOUNOUA W, BAKDI A.Fault detection and diagnosis of nonlinear dynamical processes through correlation dimension and fractal analysis based dynamic kernel PCA[J]. Chemical engineering science, 2021, 229: 116099.
[19] 薛阳, 雷文平, 岳帅旭, 等. 多模态学习方法在滚动轴承故障诊断中的应用[J]. 机械科学与技术, 2022, 41(8): 1149-1153.
XUE Y, LEI W P, YUE S X, et al.Application of multimodal deep learning method in rolling bearing fault diagnosis[J]. Mechanical science and technology for aerospace engineering, 2022, 41(8): 1149-1153.
[20] XIE T, LI Z C, WANG T Z, et al.An integration fault detection method using stator voltage for marine current turbines[J]. Ocean engineering, 2021, 226: 108808.
[21] 李恩文, 王力农, 宋斌, 等. 基于混沌序列的变压器油色谱数据并行聚类分析[J]. 电工技术学报, 2019, 34(24): 5104-5114.
LI E W, WANG L N, SONG B, et al.Parallel clustering analysis of dissolved gas analysis data based on chaotic sequences[J]. Transactions of China Electrotechnical Society, 2019, 34(24): 5104-5114.