基于聚合经验模态分解的电压闪变监测技术研究

发布时间：2018-05-11 17:58

本文选题：电压闪变 + 聚合经验模态分解　；参考：《山东理工大学》2017年硕士论文

【摘要】：随着电力电子技术和新能源技术的发展,电网中非线性负荷日益增多,造成众多电能质量问题。电压波动与闪变正是其中重要指标之一。因此研究电压波动与闪变具有十分重要意义。本文首先研究了电压波动与闪变特征参数及常用的电压闪变检测方法(平方检测法、小波分解与同步检测法、希尔伯特-黄检测法)的优点和缺点。在此基础上,本文提出了一种基于聚合经验模态分解的电压闪变检测方法,该算法首先采用希尔伯特变换提取电压闪变信号的包络,然后利用聚合经验模态分解消噪法消除包络信号中的噪声,最后采用聚合经验模态分解对消噪后包络信号进行分解,提取调幅波的特征信息。电压闪变的开始与结束时间也可以精确求出。基于MATLAB仿真电压闪变信号,并分别采用聚合经验模态检测法与常用检测法提取闪变特征信息。通过分析对比检测数据,实验结果表明,该方法可从时域与频域两方面对电压闪变信号进行分析,能够准确检测出突变、非平稳电压闪变信号的时间、幅值及频率,比小波法、希尔伯特-黄变换法等常用的闪变检测方法检测精度要高。设计了基于虚拟仪器技术的电压闪变监测系统,包括硬件部分与软件部分。硬件包括传感器、信号调理电路、数据采集卡、计算机;采用LabVIEW和MATLAB混合编程的方法实现了系统的软件设计,包括:信号定时采样程序、消噪程序、调幅波提取程序,电压参数检测程序、短时间闪变值计算程序、数据存储程序等。采用SQL Sever建立了电压闪变测量数据库,在LabVIEW中利用ADO技术访问数据库,采用LabVIEW中的ActiveX模块以及SQL语言实现了大量实时数据的存储和历史数据的查询。利用监测系统对仿真的复杂电压闪变信号进行了分析,同时系统对用高精度波形发生器产生的带噪声单频闪变信号进行了监测。最后,在智能电网研究中心的电机拖动实验室,对现场电压进行实时的在线监测,实验表明该监测系统运行稳定,可靠性高。
[Abstract]:With the development of power electronics technology and new energy technology, the nonlinear load in power grid is increasing day by day, resulting in a lot of power quality problems. Voltage fluctuation and flicker is one of the important indexes. So it is very important to study voltage fluctuation and flicker. In this paper, the advantages and disadvantages of voltage fluctuation and flicker characteristic parameters and voltage flicker detection methods (squared detection, wavelet decomposition and synchronous detection, Hilbert-yellow detection) are studied. On this basis, a voltage flicker detection method based on polymeric empirical mode decomposition is proposed. Firstly, Hilbert transform is used to extract the envelope of voltage flicker signal. Then the noise in the envelope signal is eliminated by using the polymeric empirical mode decomposition method. Finally, the envelope signal after denoising is decomposed by the polymeric empirical mode decomposition, and the characteristic information of the amplitude modulation wave is extracted. The start and end times of voltage flicker can also be accurately calculated. The voltage flicker signal is simulated based on MATLAB, and the flicker characteristic information is extracted by the polymeric empirical mode detection method and the common detection method respectively. By analyzing and comparing the detection data, the experimental results show that the method can analyze the voltage flicker signals in time domain and frequency domain, and can accurately detect the time, amplitude and frequency of the sudden and non-stationary voltage flicker signals. The accuracy of flicker detection methods such as Hilbert-Huang transform method is high. A voltage flicker monitoring system based on virtual instrument technology is designed, including hardware and software. The hardware includes sensor, signal conditioning circuit, data acquisition card, computer, and the software design of the system is realized by using LabVIEW and MATLAB mixed programming method, including signal timing sampling program, de-noising program, amplitude modulation wave extraction program. Voltage parameter detection procedures, short-time flicker calculation procedures, data storage procedures. The voltage flicker measurement database is established by SQL Sever, the database is accessed by ADO technology in LabVIEW, the ActiveX module in LabVIEW and the SQL language are used to realize the storage of a lot of real-time data and the query of historical data. The simulated complex voltage flicker signal is analyzed by using the monitoring system, and the single stroboscopic signal with noise generated by the high-precision waveform generator is monitored. Finally, the real-time on-line monitoring of the field voltage is carried out in the motor drive laboratory of the smart grid research center. The experimental results show that the monitoring system is stable and reliable.
【学位授予单位】：山东理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM933.2

【参考文献】