时变参数条件下的配电网谐波责任划分研究
发布时间：2018-10-18 19:25
【摘要】：随着电力电子技术的发展,越来越多的非线性负荷接入配电网,给电网注入大量的谐波,引起电能质量恶化。为保证供电质量的治理有据可依,必须对配电网中各谐波源的谐波责任进行合理地定量划分。在实际电网中,存在系统谐波波动、系统谐波阻抗改变、用户谐波阻抗改变等参数时变工况,导致传统的谐波责任划分方法无法满足计算准确度的要求,因此需要开展时变参数条件下的配电网谐波责任划分研究。本文研究了上述三种工况下的谐波责任定量划分方法。将主导波动量法和分位数回归法结合的方法用于划分系统谐波波动情况下谐波源的谐波责任。首先采用主导波动量法,筛选用户主导的波动量样本,消除系统谐波波动带来的影响,准确地估计系统谐波阻抗;在此基础上,将谐波责任划分问题转化为回归方程截距的求取问题,采用分位数回归法实现了系统谐波波动情况下谐波责任的准确划分,该方法稳健性强,并充分利用了系统谐波电流的波动规律。通过在IEEE 13节点系统中的仿真分析,验证了主导波动量法和分位数回归法结合的方法在系统谐波波动的工况下具有准确性高和适应性强的优点。将小波变换模极大值法和稳健整体最小二乘回归法结合的方法用于划分系统谐波阻抗改变情况下谐波源的谐波责任。首先对谐波电压和谐波电流测量数据进行加窗简化处理,得到系统谐波阻抗的粗略估值,利用小波变换模极大值法检测出系统谐波阻抗发生改变的时间,以此对测量数据进行分段处理;在此基础上,对每一段的数据采用稳健整体最小二乘回归对系统谐波阻抗进行精确估计,进而求得每段的谐波责任。最后,采用文中所定义的总谐波责任指标来定量评估谐波源的谐波责任。分别在三馈线系统和IEEE 13节点系统中进行仿真分析,验证了小波变换模极大值法和稳健整体最小二乘回归法结合的方法在系统谐波阻抗改变情况下的有效性和准确性。将变系数回归法用于划分用户谐波阻抗改变情况下谐波源的谐波责任。首先建立谐波电压和谐波电流的时变关系模型,再利用变系数回归法计算模型的回归系数,根据回归系数形成系统谐波阻抗向量,并以此计算各谐波源的总谐波责任。在IEEE 13节点系统中进行仿真分析,验证了变系数回归法在用户谐波阻抗改变情况下能够准确地估计系统谐波阻抗和谐波责任,并能有效地跟踪谐波阻抗的变化规律。论文所做的理论研究和仿真结果表明:本文提出的三种方法分别适用于系统谐波波动、系统谐波阻抗改变、用户谐波阻抗改变三种实际工况下的谐波责任定量划分,计算结果更贴近于理论值,能为未来谐波监督和谐波治理提供重要的理论依据。
[Abstract]:With the development of power electronics technology, more and more nonlinear loads are connected to the distribution network, and a large number of harmonics are injected into the power network, which results in the deterioration of power quality. In order to ensure the management of power supply quality can be relied on, it is necessary to reasonably and quantitatively divide the harmonic responsibility of each harmonic source in the distribution network. In the actual power network, there are system harmonic fluctuations, system harmonic impedance changes, user harmonic impedance changes and other parameters of time-varying operating conditions, resulting in the traditional harmonic responsibility division method can not meet the calculation accuracy requirements. Therefore, it is necessary to carry out the research on the distribution network harmonic responsibility division under the condition of time-varying parameters. In this paper, the method of quantitative division of harmonic responsibility under the above three working conditions is studied. The dominant volatility method and the quantile regression method are combined to divide the harmonic responsibility of the harmonic source in the case of harmonic fluctuation of the system. Firstly, the dominant fluctuation method is used to screen the sample of the user-led fluctuation, to eliminate the influence of the harmonic fluctuation of the system, and to estimate the harmonic impedance of the system accurately. The problem of harmonic responsibility division is transformed into the problem of obtaining the intercept of regression equation. The quantile regression method is used to realize the accurate division of harmonic responsibility in the case of harmonic fluctuation of the system. The method is robust. And make full use of the system harmonic current fluctuation law. Through simulation analysis in IEEE 13-bus system, it is verified that the combination of dominant volatility method and quantile regression method has the advantages of high accuracy and strong adaptability under the condition of harmonic fluctuation of the system. The wavelet transform modulus maximum method and the robust global least square regression method are combined to divide the harmonic responsibility of the harmonic source when the harmonic impedance of the system changes. Firstly, the harmonic voltage and harmonic current measurement data are processed by windowing, and the rough estimation of the harmonic impedance is obtained, and the time when the harmonic impedance changes is detected by the wavelet transform modulus maximum method. On the basis of this, the harmonic impedance of the system is estimated by robust global least square regression, and the harmonic responsibility of each section is obtained. Finally, the total harmonic liability index defined in this paper is used to quantitatively evaluate the harmonic responsibility of harmonic sources. The simulation analysis is carried out in the triple-fed system and the IEEE 13-bus system respectively, and the validity and accuracy of the wavelet transform modulus maximum method and the robust global least square regression method in the case of harmonic impedance change are verified. The variable coefficient regression method is used to divide the harmonic responsibility of the harmonic source when the user's harmonic impedance changes. First, the time-varying relation model of harmonic voltage and harmonic current is established, then the regression coefficient of the model is calculated by using the variable coefficient regression method, and the harmonic impedance vector of the system is formed according to the regression coefficient, and the total harmonic responsibility of each harmonic source is calculated. The simulation analysis in IEEE 13-bus system verifies that the variable coefficient regression method can accurately estimate the harmonic impedance and harmonic responsibility of the system when the user's harmonic impedance changes, and can effectively track the variation rule of harmonic impedance. The theoretical research and simulation results show that the three methods proposed in this paper are suitable for the quantitative division of harmonic responsibility under three actual operating conditions, namely, the harmonic fluctuation of the system, the change of the harmonic impedance of the system, and the change of the harmonic impedance of the user. The calculated results are closer to the theoretical values and can provide an important theoretical basis for harmonic supervision and harmonic control in the future.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711
本文编号：2280133
[Abstract]:With the development of power electronics technology, more and more nonlinear loads are connected to the distribution network, and a large number of harmonics are injected into the power network, which results in the deterioration of power quality. In order to ensure the management of power supply quality can be relied on, it is necessary to reasonably and quantitatively divide the harmonic responsibility of each harmonic source in the distribution network. In the actual power network, there are system harmonic fluctuations, system harmonic impedance changes, user harmonic impedance changes and other parameters of time-varying operating conditions, resulting in the traditional harmonic responsibility division method can not meet the calculation accuracy requirements. Therefore, it is necessary to carry out the research on the distribution network harmonic responsibility division under the condition of time-varying parameters. In this paper, the method of quantitative division of harmonic responsibility under the above three working conditions is studied. The dominant volatility method and the quantile regression method are combined to divide the harmonic responsibility of the harmonic source in the case of harmonic fluctuation of the system. Firstly, the dominant fluctuation method is used to screen the sample of the user-led fluctuation, to eliminate the influence of the harmonic fluctuation of the system, and to estimate the harmonic impedance of the system accurately. The problem of harmonic responsibility division is transformed into the problem of obtaining the intercept of regression equation. The quantile regression method is used to realize the accurate division of harmonic responsibility in the case of harmonic fluctuation of the system. The method is robust. And make full use of the system harmonic current fluctuation law. Through simulation analysis in IEEE 13-bus system, it is verified that the combination of dominant volatility method and quantile regression method has the advantages of high accuracy and strong adaptability under the condition of harmonic fluctuation of the system. The wavelet transform modulus maximum method and the robust global least square regression method are combined to divide the harmonic responsibility of the harmonic source when the harmonic impedance of the system changes. Firstly, the harmonic voltage and harmonic current measurement data are processed by windowing, and the rough estimation of the harmonic impedance is obtained, and the time when the harmonic impedance changes is detected by the wavelet transform modulus maximum method. On the basis of this, the harmonic impedance of the system is estimated by robust global least square regression, and the harmonic responsibility of each section is obtained. Finally, the total harmonic liability index defined in this paper is used to quantitatively evaluate the harmonic responsibility of harmonic sources. The simulation analysis is carried out in the triple-fed system and the IEEE 13-bus system respectively, and the validity and accuracy of the wavelet transform modulus maximum method and the robust global least square regression method in the case of harmonic impedance change are verified. The variable coefficient regression method is used to divide the harmonic responsibility of the harmonic source when the user's harmonic impedance changes. First, the time-varying relation model of harmonic voltage and harmonic current is established, then the regression coefficient of the model is calculated by using the variable coefficient regression method, and the harmonic impedance vector of the system is formed according to the regression coefficient, and the total harmonic responsibility of each harmonic source is calculated. The simulation analysis in IEEE 13-bus system verifies that the variable coefficient regression method can accurately estimate the harmonic impedance and harmonic responsibility of the system when the user's harmonic impedance changes, and can effectively track the variation rule of harmonic impedance. The theoretical research and simulation results show that the three methods proposed in this paper are suitable for the quantitative division of harmonic responsibility under three actual operating conditions, namely, the harmonic fluctuation of the system, the change of the harmonic impedance of the system, and the change of the harmonic impedance of the user. The calculated results are closer to the theoretical values and can provide an important theoretical basis for harmonic supervision and harmonic control in the future.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM711
【相似文献】
相关期刊论文 前10条
1 康婕;解绍锋;刘晓菊;魏晓娟;;基于支持向量机的谐波阻抗估计方法[J];电力系统保护与控制;2010年22期
2 王诗超;沈沉;程建洲;;考虑电流波动特性的系统侧谐波阻抗估计方法[J];电力系统自动化;2012年03期
3 蔡景青;;谐波阻抗估计的方法及应用研究[J];电子技术与软件工程;2013年22期
4 苏攀;郭成;易东;;基于中位参数等价权回归法的谐波阻抗分析[J];云南电力技术;2014年01期
5 蔡景青;;谐波阻抗估计方法及应用[J];农村电气化;2013年S1期
6 华回春;贾秀芳;曹东升;赵成勇;;系统谐波阻抗估计的极大似然估计方法[J];中国电机工程学报;2014年10期
7 张安安,杨洪耕;基于畸变波形同步分层估计谐波阻抗的探讨[J];电力系统自动化;2003年09期
8 李群湛,王彦东,刘学,陈仕龙,董志杰;基于稳态测量的电力网络谐波阻抗的确定[J];电工技术杂志;2004年07期
9 刘秀玲;王洪瑞;;电网谐波阻抗的测量及其修正[J];电力系统及其自动化学报;2007年02期
10 丰卫平;电力系统谐波阻抗的测量计算法[J];云南电力技术;1999年S1期
相关会议论文 前2条
1 王磊;;基于“分组线性回归法”测量谐波阻抗[A];中国高等学校电力系统及其自动化专业第二十四届学术年会论文集（下册）[C];2008年
2 师洪涛;卓放;侯李祥;岳小龙;张东;耿志清;;三相PWM变流器谐波阻抗模型及测量研究[A];分布式发电、智能微电网与电能质量——第三届全国电能质量学术会议暨电能质量行业发展论坛论文集[C];2013年
相关博士学位论文 前1条
1 李飞;分布式发电并网逆变器的虚拟谐波阻抗控制研究[D];合肥工业大学;2015年
相关硕士学位论文 前10条
1 徐玉刚;基于谐波阻抗测量技术的配电网三相自适应重合闸技术研究[D];山东大学;2015年
2 韩劲草;有源电力滤波器的谐波检测和谐波阻抗估计[D];北京理工大学;2016年
3 姜昊;电网阻频特性测试与辨识[D];西南交通大学;2016年
4 刘哲;典型干扰性负荷建模及其谐波污染责任计算[D];华北电力大学;2016年
5 陈静;时变参数条件下的配电网谐波责任划分研究[D];西南交通大学;2017年
6 张安安;基于畸变波形同步分层估计谐波阻抗[D];四川大学;2003年
7 王彦东;电网谐波阻抗特性及测量方法研究[D];西南交通大学;2004年
8 毕会静;供电系统谐波辨识的研究[D];华北电力大学（河北）;2006年
9 苏攀;牵引变电所阻频特性的辨识研究[D];西南交通大学;2014年
10 马U_潇;高速电气化铁路谐振故障的分析与研究[D];西南交通大学;2012年
，本文编号：2280133
本文链接：https://www.wllwen.com/kejilunwen/dianlidianqilunwen/2280133.html
