【作者】 李飞；

【导师】 张兴；

【作者基本信息】 合肥工业大学 ， 电气工程， 2015， 博士

【摘要】 随着分布式发电的大规模应用和电网电能质量问题的日益严峻,电网迫切需要分布式发电并网逆变器在实现传输可再生能源发出的功率同时,提供必要的谐波抑制功能。诸多学者研究了具有谐波补偿控制的功能复用并网逆变器,即将传统的分布式发电并网逆变器,功能复用,在传输有功功率的同时,实现APF功能,来补偿电网谐波。然而,实际应用中,由于并网逆变器难以获得负载谐波电流信息,从而无法补偿谐波。并且该方法无法补偿电网电压背景谐波造成的PCC电压畸变。在此背景下,本文研究了分布式发电并网逆变器的虚拟谐波阻抗控制。相对于并网逆变器的谐波补偿控制,虚拟谐波阻抗控制只需要从容易获得的公共连接点(Point of Common Coupling, PCC)电压信号来提取谐波信息,不受谐波电流检测的限制,并且对电网背景谐波带来的PCC电压畸变也有所抑制。本文以并网逆变器的谐波抑制功能为起点,深入地研究了分布式发电并网逆变器虚拟谐波阻抗控制的若干问题,获得了一些具有创新性的研究成果。1、针对谐波阻抗对谐波抑制的效果缺少定量分析的问题,提出了一种基于谐波抑制系数的可视化虚拟谐波阻抗抑制效果分析方法。已有文献,主要集中于虚拟谐波阻抗的定性分析,缺少对其谐波抑制效果的定量研究。本文定义了谐波抑制系数的概念,可以同时反映谐波阻抗对电网电压和谐波负载带来的PCC电压和电网电流畸变的抑制能力,因而十分适宜用来定量评价谐波阻抗抑制能力。在此基础上,提出一种可视化谐波阻抗抑制能力分析方法,该方法使用三维立体图像,可以形象直观的展现出虚拟谐波阻抗的谐波抑制能力随谐波阻抗参数的变化趋势。2、针对传统方法中电流环控制器参数对虚拟谐波阻抗稳定性分析的影响,研究了一种基于D分割法的虚拟谐波阻抗稳定性分析方法。传统的稳定性分析方法,诸如波特图、零极点图或根轨迹,都需要确定系统的所有参数。使用传统方法来分析谐波阻抗值对系统稳定性造成的影响时,内部电流跟踪控制器参数的不恰当选择,不可避免的会造成稳定性分析结果的误差,本文采用D分割法对虚拟谐波阻抗进行稳定性分析。该方法通过比较电流环控制器参数稳定域面积来分析谐波阻抗对系统稳定性的影响,不依赖具体控制器参数值,可以反映虚拟谐波阻抗算法本身的稳定性。3、针对电阻型虚拟谐波阻抗的谐波抑制效果与稳定性相互矛盾的问题,提出并研究了一种阻感型虚拟谐波阻抗控制方法。已有的文献对于虚拟谐波阻抗的研究,都将谐波阻抗直接定义为纯电阻特性。然而,纯电阻特性不一定为最佳的阻抗形式。本文通过理论推导得出最佳阻抗的数学表达式,发现阻感型虚拟谐波阻抗相对于电阻型虚拟谐波阻抗,具有更强的低频谐波抑制能力和系统稳定性。在此基础上提出一种并网逆变器的阻感型虚拟谐波阻抗控制,并对其谐波抑制效果和稳定性进行了分析。4、针对阻感型虚拟谐波阻抗存在的高频谐波放大问题,提出并研究了一种复合型虚拟谐波阻抗控制方法。阻感型虚拟谐波阻抗控制方法会带来高频谐波的放大,针对这一问题,进一步提出了一种复合型虚拟谐波阻抗控制方法。该复合型虚拟谐波阻抗由电阻型虚拟谐波阻抗和阻感型虚拟谐波阻抗并联构成。在低频段,复合型虚拟谐波阻抗主要呈阻感特性,具有较好的谐波抑制效果,在高频段,主要呈电阻特性,避免了谐波放大的发生。此外,复合型虚拟谐波阻抗相对于电阻型虚拟谐波阻抗稳定性更强。5、针对功能复用并网逆变器对滤波器带宽的特殊要求,提出了一种LCL-LC滤波器拓扑及其参数设计方法。为了提高滤波器带宽,文章研究了滤波器的拓扑演变过程,在此基础上,提出了一种LCL-LC滤波器拓扑。所提的LCL-LC滤波器拓扑,由传统的LCL滤波器与一个LC串联谐振支路共同构成。由于LCL-LC滤波器的串联谐振支路能有效旁路开关频率谐波,滤波器可以以频率较高幅值较小的二倍开关频率谐波为滤除目标,故该滤波器相对于传统的LCL滤波器具有较高的带宽优势,同时也降低了滤波器的成本。并且,本文分析并得出了所提滤波器的两个谐振频率特性,在此基础上提出了一种LCL-LC滤波器的参数设计方法。此外、针对所提滤波器参数个数较多,传统的参数鲁棒性分析方法无法适用的问题,还进一步提出了一种可视化参数鲁棒性校验方法。与传统方法相比,所提的方法可以在无需反复迭代的情况下同时校验多个参数发生偏移的情况,并可以清楚的展现参数变化对滤波器性能影响的变化趋势。6、针对虚拟谐波阻抗控制逆变器对带宽的高要求,研究了一种基于D分割法的高带宽电流环PI调节器参数设计方法。首先将连续域D分割法扩展到离散域,在此基础上提出了一种基于高带宽的电流环PI调节器参数设计方法。该方法利用D分割法得到满足设定的幅值裕度和相位裕度的PI调节器参数稳定域,在此参数域内,以高带宽为目标函数,选择PI调节器具体参数值。更多还原

【Abstract】 With the large-scale application of the distributed power generation (DG) and growing severe grid power issues, the grid connected inverter in DG is required to provide the necessary harmonic suppression services, in addition to transfer power from DG to grid. The multi-functional grid-connected inverter(MFGCI) in DG with harmonic compensation control, which means a grid-connected inverter in DG is combined with active power filter(APF) services and can compensate harmonics in grid, has been garnered a lot of attention. But, it is very difficult to obtain the harmonic currents information of distant loads to compensate them and it can’t compensate the voltage distortion at Point of Common Coupling (PCC) caused by background harmonic voltage in power systems. Aiming at this issue, the virtual harmonic impedance control for MFGCI is researched in this thesis. Compared with the harmonic compensation control of grid connected inverter, the virtual harmonic impedance control can not only compensate harmonic currents without the limitation of harmonic current detection, but also suppress PCC voltage harmonic caused by background harmonic voltage. Face to the increasingly desired ancillary services of grid-connected inverter in DG, some detailed and comprehensive studies on the virtual harmonic impedance control for MFGCI have highlighted in this dissertation thesis. And some novel and important conclusions are drawn as follows.1. Aiming at quantitatively evaluating harmonic suppression effect of the harmonic impedance, A harmonic suppression coefficient based visual method to analyze the suppressing ability of virtual harmonic impedances is proposed. Literatures have mainly focused on the qualitative analysis of resistive harmonic impedance, and the lack of a quantitative study of its harmonic suppression effect. This thesis proposes a concept of harmonic suppression coefficient. The coefficient can show suppressing ability of the virtual harmonic impedance to PCC voltage and grid current distortion which is caused by grid background harmonic voltage or harmonic loads. It makes the coefficient is very suitable as a criteria to evaluate the effect of virtual harmonic impedances. On this basis, a visual method to analyze the suppressing ability of virtual harmonic impedances is proposed. The proposed method can visually show variation trends of virtual harmonic impedances suppressing ability by a three-dimensional image.2. A D-partition based virtual harmonic impedance stability analysis method is researched in order to avoid the influence of current controller parameters in harmonic impedance stability analysis. In the traditional stability analysis methods, such as bode plots, zero pole map or root locus, all the parameters of system are necessary. And the different current controller parameters will inevitably affect the result of stability analysis. So, a D-decomposition based virtual harmonic impedance stability analysis method is proposed in the thesis to avoid the influence of current controller parameters. The method can show the stability of virtual harmonic impedance by comparing the stability region area of current controller parameters and does not need specific controller parameter value.3. Aiming at the conflict of harmonic suppression effect and stability in Resistive harmonic impedance control, the Resistive and Inductive (RL) harmonic impedance control strategy for MFGCI is proposed in this thesis. The harmonic impedance in the most of literature is resistive harmonic impedance. But, is the resistive the best form of harmonic impedance? For this problem, the system is modeled and the mathematical expression of optimal harmonic impedance is derived. From this, a resistive and inductive harmonic impedance control strategy is proposed which has better harmonic suppression ability at low frequency and system stability.4. In order to avoid the high frequency harmonic amplification in the RL harmonic impedance control, a composite harmonic impedance control strategy for MFGCI is proposed in this thesis. Aiming at the problem of high frequency harmonic amplification due to resistive and inductive harmonic impedance control strategy, composite harmonic impedance is proposed. The composite harmonic impedance is composed by resistive harmonic impedance and resistive and inductive harmonic impedance in parallel. The composite harmonic impedance has a better harmonic suppression effect at low frequency and no harmonic amplification at high frequency. In addition, the composite harmonic impedance is more stable then resistive harmonic impedance.5. As to the bandwidth requirement of out put filter in MFGCI, an LCL-LC filter topology and its parameter design are presented in the thesis. In order to improve the bandwidth of output filter, the evolution of filter is studied, and an LCL-LC filter topology is proposed, which is a combination of a LCL filter and a LC series resonant circuit. The filter has the advantages of high bandwidth and low cost compared with the traditional LCL filter. Two resonant frequency characteristics of the filter are analyzed and a parameter design method on the base of the characteristics is also proposed in the paper.On the other hand, the number of parameters in presented LCL-LC filter is too big to analyze its parameter robustness by the traditional way. And then a parameter robustness analysis method based on four-dimensional graphics is proposed to analyze parameter robustness of the presented filter. Compared with the traditional ones, the proposed analysis method can analyze the filter performance under variations of several parameters at a time without any iteration and show the trend of filter performance due to filter parameters variation.6. According to the requirement of bandwidth in MFGCI, A high bandwidth PI controller parameters design on the base of D-partition method is researched. Firstly, the continuous domain D-partition method is extended to the discrete domain, and on this basis, the author proposes a PI controller parameters design method with high bandwidth. The method calculates the stability region under the setting magnitude margin and phase margin by discrete D-partition method, and then choose the select the PI parameters with the highest bandwidth.更多还原