【作者】 刘昊；

【导师】 唐轶；

【作者基本信息】 中国矿业大学 ， 电力电子与电力传动， 2009， 博士

【摘要】 随着时代的进步与科技的发展,越来越多的大功率电力电子设备被使用在电力系统中,产生诸如谐波,电压暂降等扰动,导致电能质量问题愈加突出。另一方面,电力系统中的电能质量问题又会影响一些电力电子设备的安全运行。如何正确地识别电能质量扰动,并对扰动程度进行评估已成为当下亟待解决的问题。这也是以后控制及治理电能质量问题的基础。基于此,本文着重于研究电能质量扰动分类及参数评估问题,主要贡献如下:在分析国内外现有电能质量扰动分类方法的基础上,研究了常见扰动数学表达式,并将常见电压扰动分为电压幅值扰动和加性扰动。即一类为对基波幅值产生的扰动,一类为在基波信号上叠加一个扰动信号的扰动。又根据扰动时间长短,将每一大类分为暂态和稳态扰动。基于此分类,针对不同类型的扰动特点,提取了四个特征量F1~F4, F1反映了基波幅值的变化,可用于判断电压幅值扰动(如电压暂升、暂降、中断及闪变);F2~F4可被用来判断加性扰动(如谐波,间谐波,振荡暂态等)。F2等同于总谐波畸变率, F3表明基波幅值U1f曲线的平均变化量,F4表明U1f曲线的突变量。所提取的扰动分类特征量有以下特点:(1)某一特征量或某几个特征量能唯一地确定一种扰动,使扰动判断不再具有或然性;(2)任一扰动的特征量不因其它扰动的存在而变化较大,因此可以准确辨别出混合扰动;(3)具有较为明确的物理意义,因此有利于对各种电能质量扰动进行估值。针对所提取特征量的特点,本文提出一种基于时域特性分析的电能质量扰动分类方法。该方法用基于规则的分类方法对常见电能质量扰动进行了分类。MATLAB仿真和分析表明,该方法能有效地识别除切痕外所有基本和混合电能质量扰动,包括三种和四种扰动同时存在的情况,并且计算耗时少,适合在线分析。在此分类方法的基础上,对常见电能质量扰动进行了参数评估。因为提取的四个时域特性参数都有较明确的物理意义,故较易对扰动参数进行评估。F 1可用来评估电压暂升暂降发生的起止时间及幅度,还可用来评估闪变调制波的频率和幅值,另外,U1f曲线还可用来拟合振荡暂态的各个参数如振荡衰减常数,振荡频率等; F2可用来精确评估总谐波畸变率这一常用电能质量参数。为验证理论的正确性,作者研发了电能质量分析仪样机。分析仪的软件算法以所提出扰动分类方法为理论依据。实验采用dSPACE公司生产的半实物仿真平台作为扰动信号发生器对电能质量分析仪的扰动识别效果进行了检测,并进行了实验室模拟实验及现场在线测试实验。1000次半实物仿真实验,20次的实验室模拟实验,20次的现场在线测量结果表明,所研究算法理论正确,分类方法有效可行,参数估计基本能够反映扰动对电力系统的影响程度,研发的电能质量分析仪可较好地对常见电能质量扰动进行识别并对扰动参数进行较准确的评估。更多还原

【Abstract】 With the development of the science and technology, more and more nonlinear and impactive load are used in the power system. And it makes the power quality problems more and more serious. The power quality disturbances like voltage sag, swell, fluctuation and flicker, and oscillatory transients appear in the power system more frequently. So the traditional evaluation method like the index power supply reliability is not enough for the demand of the power quality evaluation nowadays. It is a desiderating problem to be solved that how to classify the power quality disturbances correctly and evaluate the disturbance parameters. It is also the basis of the control and improvement of the power quality problems.In detail, the major contributions of the thesis are as following:A novel classification method of power quality disturbances is proposed in the paper on the basis of the analysis of present power quality disturbance classification methods in and abroad. Four characteristic parameters F1~F4 are proposed based on analyzing the power quality disturbances signals time domain characteristics. And the parameters can be used to classify power quality disturbances well. F1 reflects the variation of the fundamental wave amplitude. It can be used to judge whether the voltage amplitude disturbances exist (like voltage sag, swell, interruption and fluctuation). F2~F4 can be used to judge whether additive disturbances exist (like harmonics, interharmonics, oscillatory transients). F2 equates the THD (Total Harmonic Distortion). F3 indicates the average variation of F1. And F4 indicates the mutation of F1. The features extracted by this method have advantages as follows: (1) The features have clear physical meanings. So it profits to evaluate the disturbance parameters. (2) Some feature or some features can determine one disturbance only. Then the disturbance classification will not be probability. (3) The features of some disturbance will not change because of other disturbances existence. So it can classify mixed disturbances. Matlab simulation and analysis show that this proposed method can classify more single or complex disturbances signals effectively and loses less computation time of detecting and classifying electrical disturbances. So it is proper to be used in online analysis.The evaluation of the common power quality disturbances are made based on the classification method proposed. It is easy to evaluate because the four characteristic parameters all have clear physical meanings. F1 can be used to evaluate the starting time, the end time and the amplitude of the voltage sag and swell. It can also be used to evaluate the fluctuation and flicker’s modulation frequency and amplitude. Otherwise, the curve of F1 can be used to fit the oscillatory transients attenuation constant and the oscillatory frequency. F2 can be used to evaluate frequently used power quality index—the voltage harmonics content.The author develops a power quality analyzer based on TMS320F2812 DSP to prove the theory. The software of the analyzer is the algorithm mentioned above. TMS320F2812 of TI Company is used as the CPU to assure the real-time calculation. And the peripheral hardware includes 16 bit A/D converter microchip AD7656 which can assure the accuracy of the A/D. The experiment tests the classification effect of the analyzer using dSPACE DS1104 A/D D/A converter board of dSPACE Company as the disturbance signals generator. The results shows that the power quality analyzer developed can classify the power quality disturbances correctly and evaluate the disturbance index accurately.更多还原