柔性直流输变电的故障检测技术
|
摘要
柔性直流输变电系统在输电配电过程中受到负载干扰等因素的影响,容易产生柔性直流输变电故障,为了提高故障诊断效率,提出一种基于输变电异常特征提取和自适应神经网络的柔性直流输变电的故障检测技术。采用多传感器量化融合方法进行柔性直流输变电系统的关联信息采集,提取柔性直流输变电的多维信息特征量,采用相关性波束分析方法进行柔性直流输变电的频谱分解和关联规则挖掘,结合谱分析方法进行输变电的异常特征提取,根据频谱差异性进行柔性直流输变电故障判断和故障类型辨识,对提取的柔性直流输变电系统故障特征量采用神经网络学习算法进行分类识别,实现柔性直流输变电故障优化检测。仿真结果表明,采用该方法进行柔性直流输变电故障检测的准确性较高,抗干扰能力较强,提高了输变电的故障检测能力。
The flexible DC transmission and transformation system is affected by the load interference in the transmission and distribution process,and it is easy to produce flexible DC transmission and transformation fault.In order to improve the efficiency of fault diagnosis,a fault detection technique based on the characteristic extraction of transmission and transformation and adaptive neural network is proposed.The multi-sensor quantization fusion method is used to collect the correlation information of the flexible DC transmission and transformation system,and the multidimensional information characteristic quantity of the flexible DC transmission and transformation system is extracted.Correlation beam analysis method is used to decompose the spectrum and mining association rules of flexible DC power transmission and transformation,and the abnormal features of transmission and transformation are extracted by combining spectrum analysis method.According to the difference of frequency spectrum,the fault diagnosis and fault type identification of flexible DC transmission and transformation system are carried out,and the fault characteristics of flexible DC transmission and transformation system are classified and identified by neural network learning algorithm.Realize the flexible DC transmission and transformation fault optimization detection.The simulation results show that the proposed method has high accuracy and strong anti-interference ability for flexible DC transmission and transformation fault detection,and the fault detection ability of transmission and transformation is improved.
The flexible DC transmission and transformation system is affected by the load interference in the transmission and distribution process,and it is easy to produce flexible DC transmission and transformation fault.In order to improve the efficiency of fault diagnosis,a fault detection technique based on the characteristic extraction of transmission and transformation and adaptive neural network is proposed.The multi-sensor quantization fusion method is used to collect the correlation information of the flexible DC transmission and transformation system,and the multidimensional information characteristic quantity of the flexible DC transmission and transformation system is extracted.Correlation beam analysis method is used to decompose the spectrum and mining association rules of flexible DC power transmission and transformation,and the abnormal features of transmission and transformation are extracted by combining spectrum analysis method.According to the difference of frequency spectrum,the fault diagnosis and fault type identification of flexible DC transmission and transformation system are carried out,and the fault characteristics of flexible DC transmission and transformation system are classified and identified by neural network learning algorithm.Realize the flexible DC transmission and transformation fault optimization detection.The simulation results show that the proposed method has high accuracy and strong anti-interference ability for flexible DC transmission and transformation fault detection,and the fault detection ability of transmission and transformation is improved.
引文
[1] 王海峰,陆军.复杂电力设备突发故障诊断方法研究与仿真[J].计算机仿真,2013,30(3):127-129.
[2] 张宏利,刘吉成,李栋,等.基于免疫机理的设备异常度检测与故障诊断快速融合方法研究[J].机械设计与制造工程2016,45(11):96-101.
[3] 戴金玲,许爱强.一种变异真值表故障模型的神经网络测试生成算法[J].电子设计工程,2017,25(19):174-178.
[4] GUO Huaping,ZHOU Jun,WU Chang’an,et al.k-nearest neighbor classification method for class-imbalanced problem[J].Journal of Computer Applications,2018,38(4):955-959.
[5] LU C,SHENG W,HAN Y,et al.Phase-only pattern synthesis based on gradient-descent optimization[J].Journal of Systems Engineering and Electronics,2016,27(2):297-307.
[6] ROEDLER R,KERGL D,RODOSEK G D.Profile matching across online social networks based on Geo-tags[C]//NaBIC 2015:Proceedings of the 7th World Congress on Nature and Biologically Inspired Computing,AISC 419.Berlin:Springer-Verlag,2016:417-428.
[7] 高闪,梅劲松.输入非线性系统的执行器故障容错控制[J].信息与控制,2015,44(4):463-468.
[8] 孙海亮,訾艳阳,何正嘉.多小波自适应分块阈值降噪及其在轧机齿轮故障诊断中的应用[J].振动工程学报,2013,26(1):127-134.
[9] 孙云嵩,于德介,陈向民,等.基于信号共振稀疏分解的阶比分析及其在齿轮故障诊断中的应用[J].振动与冲击,2013,32(16):88-94.
[10] 张世荣,林虞.结合小波去噪的PCA工业过程传感器故障检测[J].自动化与仪表,2016,31(5):1-5.
[11] ZHENG J F,ZHANG J,ZHU K Y,et al.Gust front statistical characteristics and automatic identification algorithm for CINRAD[J].Acta Meteorologica Sinica,2014,28(4):607-623.
[12] HWANG Y,YU T Y,LAKSHMANAN V,et al.Neuro-fuzzy gust front detection algorithm with S-band polarimetric radar[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(3):1618-1628.
[13] 王海斌,王永虎,丁发军,等.基于嵌入式系统的放电故障紫外检测平台[J].自动化与仪表,2016,31(7):66-69.
[14] 徐政,刘高任,张哲任.柔性直流输电网的故障保护原理研究[J].高电压技术,2017,43(1):1-8.
[15] 陈朋,李梅航,严兵,等.适用于多端柔性直流输电系统的通用控制策略[J].电力系统自动化,2016,40(21):47-52.
[2] 张宏利,刘吉成,李栋,等.基于免疫机理的设备异常度检测与故障诊断快速融合方法研究[J].机械设计与制造工程2016,45(11):96-101.
[3] 戴金玲,许爱强.一种变异真值表故障模型的神经网络测试生成算法[J].电子设计工程,2017,25(19):174-178.
[4] GUO Huaping,ZHOU Jun,WU Chang’an,et al.k-nearest neighbor classification method for class-imbalanced problem[J].Journal of Computer Applications,2018,38(4):955-959.
[5] LU C,SHENG W,HAN Y,et al.Phase-only pattern synthesis based on gradient-descent optimization[J].Journal of Systems Engineering and Electronics,2016,27(2):297-307.
[6] ROEDLER R,KERGL D,RODOSEK G D.Profile matching across online social networks based on Geo-tags[C]//NaBIC 2015:Proceedings of the 7th World Congress on Nature and Biologically Inspired Computing,AISC 419.Berlin:Springer-Verlag,2016:417-428.
[7] 高闪,梅劲松.输入非线性系统的执行器故障容错控制[J].信息与控制,2015,44(4):463-468.
[8] 孙海亮,訾艳阳,何正嘉.多小波自适应分块阈值降噪及其在轧机齿轮故障诊断中的应用[J].振动工程学报,2013,26(1):127-134.
[9] 孙云嵩,于德介,陈向民,等.基于信号共振稀疏分解的阶比分析及其在齿轮故障诊断中的应用[J].振动与冲击,2013,32(16):88-94.
[10] 张世荣,林虞.结合小波去噪的PCA工业过程传感器故障检测[J].自动化与仪表,2016,31(5):1-5.
[11] ZHENG J F,ZHANG J,ZHU K Y,et al.Gust front statistical characteristics and automatic identification algorithm for CINRAD[J].Acta Meteorologica Sinica,2014,28(4):607-623.
[12] HWANG Y,YU T Y,LAKSHMANAN V,et al.Neuro-fuzzy gust front detection algorithm with S-band polarimetric radar[J].IEEE Transactions on Geoscience and Remote Sensing,2017,55(3):1618-1628.
[13] 王海斌,王永虎,丁发军,等.基于嵌入式系统的放电故障紫外检测平台[J].自动化与仪表,2016,31(7):66-69.
[14] 徐政,刘高任,张哲任.柔性直流输电网的故障保护原理研究[J].高电压技术,2017,43(1):1-8.
[15] 陈朋,李梅航,严兵,等.适用于多端柔性直流输电系统的通用控制策略[J].电力系统自动化,2016,40(21):47-52.