作者：刘骁眸1, 马旭1, 金石炜1, 闫孝姮2, 陈宏强2
作者单位：1. 国网辽宁省电力有限公司超高压分公司, 辽宁 沈阳 110003;
2. 辽宁工程技术大学电气与控制工程学院, 辽宁 葫芦岛 125105
关键词：均压电极结垢检测;主成分分析法;改进天牛须算法;支持向量机
摘要：
针对特高压换流站传统人工定期巡检均压电极结垢状况方法落后的问题，提出基于互补集合经验模态分解(ceemd)和改进天牛须算法优化支持向量机的均压电极结垢智能检测方法。首先，模拟±500 kv特高压换流站换流阀冷却系统，开展均压电极结垢厚度为0.1~0.8 mm实验，利用超声波时域反射技术获取均压电极结垢回波信号，并采用ceemd结合小波阈值(ceemd-wt)方法对信号进行降噪预处理；然后，采用主成分分析方法(pca)筛选出主要特征向量，通过引入粒子群信息共享思想和遗传算法选择、交叉、变异对天牛须算法进行改进，并利用改进算法优化svm参数；最后，采用ibas-svm模型对均压电极结垢厚度进行检测识别。结果表明：所提方法对均压电极结垢厚度识别准确率达到91.75%，能够测出0.1~0.8 mm范围内的均压电极结垢厚度，可为特高压换流站均压电极非接触式结垢探测提供一种行之有效的方法。
sediment identification method for grading electrodes in hvdc converter station based on ceemd and ibas-svm
liu xiaomou1, ma xu1, jin shiwei1, yan xiaoheng2, chen hongqiang2
1. super high voltage branch of state grid liaoning electric power co., ltd., shenyang 110003, china;
2. college of electrical and control engineering, liaoning technical university, huludao 125105, china
abstract: addressing the question that the traditional method of the manual or periodic inspection of the grading electrode sediment status of the hvdc converter station is backward, an intelligent detection method based on complementary ensemble empirical mode decomposition(ceemd) and improved beetle antennae search to optimize support vector machine was proposed. firstly, the cooling system of converter valve in ±500 kv hvdc converter station was simulated, and the experiments of grading electrode sediment thickness of 0.1-0.8 mm were carried out. the ultrasonic time domain reflection technology was used to obtain the sediment echo signals of the grading electrode, and the signals were preprocessed by ceemd combined with wavelet threshold ( ceemd-wt ) method. then principal component analysis (pca) was used to filter out the main feature vectors, and the beetle antennae search algorithm was improved by introducing the idea of particle swarm information sharing and genetic algorithm selection, crossover and variation, and the svm parameters were optimized using the improved algorithm. finally, the ibas-svm model was used to detect and identify the thickness of the grading electrode sediment. the results show that the proposed method has an accuracy of 91.75% for the identification of the thickness of the grading electrode sediment, and can measure the thickness of the grading electrode sediment in the range of 0.1-0.8 mm, which provides an effective method for the non-contact sediment detection of the grading electrode in hvdc converter stations.
keywords: scaling detection of equalizing electrode;principal component analysis;improved beetle antennae search;support vector machine
2023, 49(6):172-180 收稿日期: 2021-09-29;收到修改稿日期: 2021-11-12
基金项目: 国家电网公司总部科技项目资助(sglnjx00yjjs2100620)
作者简介: 刘骁眸(1982-),男,河北衡水市人,高级工程师,研究方向为直流检修运维
参考文献
[1] xu z, chen h r. overview of voltage source converter type hvdc transmission technology[j]. high voltage technology, 2007, 33(1): 1-10
[2] 梁秉岗, 王晨星, 张朝辉, 等. 换流阀内冷却系统均压电极结垢分布规律[j]. 南方电网技术, 2021, 15(8): 55-63
[3] 刘学忠, 王晨星, 刘宁, 等. 换流阀内水冷系统泄漏电流对散热器腐蚀的影响[j]. 高电压技术, 2020, 46(5): 1781-1790
[4] 王恭, 罗博文, 曹生现, 等. 换热管道污垢热阻直接检测方法与实现[j]. 中国电机工程学报, 2018, 38(14): 4165-4172, 4322
[5] 王彤彤, 孙灵芳. 基于改进cosamp的管道污垢超声检测信号的压缩重构研究[j]. 机械强度, 2020, 42(6): 1316-1322
[6] 孙灵芳, 王彤彤, 徐曼菲, 等. 基于改进ceemd的薄层污垢超声检测信号去噪[j]. 仪器仪表学报, 2017, 38(12): 2879-2887
[7] oliveira m a, simas filho e f, albuquerque m c s, et al. ultrasound-based identification of damage in wind turbine blades using novelty detection[j]. ultrasonics, 2020, 108: 106166
[8] 周澄, 邓菲, 刘尧, 等. 基于神经网络和支持向量机的导波弯管腐蚀损伤程度辨识研究[j]. 机械工程学报, 2021, 57(12): 136-144
[9] 但长林, 李三雁, 张彬. 基于样本熵和svm的滚动轴承故障诊断方法研究[j]. 中国测试, 2020, 46(11): 37-42
[10] 李爱莲, 郭志斌, 解韶峰, 等. 蚁群和粒子群混合优化svm的钢板表面缺陷分类研究[j]. 中国测试, 2020, 46(1): 110-116
[11] 樊浩, 李兴文, 苏海博, 等. 基于主成分分析——支持向量机优化模型的断路器故障诊断方法研究[j]. 高压电器, 2020, 56(6): 143-151
[12] 谢国民, 倪乐水. 基于iabc优化svm的变压器故障诊断[j]. 电力系统保护与控制, 2020, 48(15): 156-163
[13] 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
[14] tong y, li j, xu y, et al. signal denoising method based on improved wavelet threshold function for microchip electrophoresis c4d equipment[j]. complexity, 2020, 2020(3): 1-11
[15] 李黄曼, 张勇, 张瑶. 基于issa优化svm的变压器故障诊断研究[j]. 电子测量与仪器学报, 2021, 35(3): 123-129
