【作者】 李幸；

【导师】 粟梅；

【作者基本信息】 中南大学 ， 控制科学与工程， 2014， 博士

【摘要】 摘要：矩阵变换器具备能量双向流动、正弦输入输出电流、输入功率因数可控、无需储能环节等一系列优良的电气特性,因而在变频调速、可再生能源发电和电能质量控制等领域具有广阔的应用前景。在一些特殊应用领域,除要求矩阵变换器产生高质量的输入电流、输出电压以外,还希望它能在最大范围内自由调节输入无功功率。然而,因结构特性等原因,矩阵变换器的输入无功与调制策略、电压传输比及负载性质等的关系较复杂,不可像双PWM变换器那样可独立调节。因此,本文主要对矩阵变换器的输入无功特性及控制问题展开深入研究。对于传统矩阵变换器拓扑,论文分析了几种经典调制算法的输入无功特性及相关约束条件,发现它们的输入无功控制范围均有不同程度的受限。对此,提出了一种基于数学构造思想的扩展无功调制策略。该策略通过构造一种含负载电流相位信息的无功调制矩阵,达到减弱输入无功功率对负载功率因数依赖的目的。相对经典调制算法,所提算法具有实现简单和输入无功范围大等优点。对于双级矩阵变换器拓扑,由于要求其中间直流电压须始终为正,因而它在空间矢量调制框架下的输入无功控制范围相对传统矩阵变换器拓扑更为受限。对此,根据等效原则,提出了一种整流、逆变同时采用‘相反矢量”代替原矢量的空间矢量合成方法(矢量倒置法),从而在一定程度上弥补了双级矩阵变换器的拓扑限制。虽然基于数学构造思想的扩展无功调制策略能大幅提高矩阵变换器的输入无功控制范围,但它无法直接应用于双级矩阵变换器。为此,提出了一种基于空间矢量合成的扩展无功调制策略。其基本思路是：首先对输出电压矢量和输入无功电流矢量进行独立空间矢量合成,再采用矢量倒置思想将上述两个合成过程进行“矢量合并”。相对于“脉冲合并”的调制策略,所提方法具有便于理解、计算量小以及输入无功控制范围更宽等特点。在双级矩阵变换器拓扑结构的基础上,提出了一种集成静止同步补偿器(STATCOM)功能的矩阵变换器新拓扑。论文阐述了该拓扑的结构特点及工作机理,并在此基础上提出了相应的调制算法,还分析了该拓扑在所提调制算法下的输入无功特性。理论上,该拓扑消除了矩阵变换器输入无功对输出电流的依赖,为矩阵变换器实现大范围输入无功控制提供了新思路。基于矩阵变换器的双馈风力发电系统的低电压跨失控制问题,实质也是矩阵变换器的无功控制问题。针对该问题,本文提出了一种基于矩阵变换器拓扑结构重构的解决方案。通过在双级矩阵变换器的直流环节中级联辅助开关电路,实现了双级矩阵变换器拓扑的结构重构。其工作原理是：电网电压正常时,辅助开关电路不工作；电网电压跌落时,立即通过触发辅助电路开关实现拓扑结构重组,即双级矩阵变换器被分解成功能上相互分离的电流型STATCOM和普通电压型逆变器。如此,该方案一方面向电网提供所需无功以支撑电网电压,另一方面能将双馈感应发电机的转子电流控制在合理水平,以确保矩阵变换器的安全运行。更多还原

【Abstract】 Abstract:Matrix converter is a kind of green power electronic converter without intermediate energy storage link, featured by sinusoidal input and output currents, bidirectional energy flow, controllable input power factor as well as a compact design, and it has found many applications in adjustable speed drives, renewable energy generation, power quality control, and so on. In some applications, expect for the desired high-quality output voltages, matrix converter is also required to obtain the input reactive power freely as much as possible. However, due to the lack of the energy storage link, the input and output sides of matrix converter would interfere with each other, the input reactive power cannot be regulated independently as that of PWM converter. Besides, the reactive power characteristics and control methods are very complex, which are related to modulation algorithm, voltage transfer ratio, and load properties and so on. Hence, this paper will focus on the characteristics and control issues of the input reactive power for matrix converter. The main works may include:For the traditional matrix converter topology, the reactive power features under several conventional modulation methods are studied, and the related control ranges and constraints are analyzed and compared, which show that the control ranges of input reactive power are limited. Thus, a reactive power-extended modulation method based on mathematic construction is proposed, which eliminates the dependence of reactive power on the load power factor by constructing a reactive modulation matrix. The proposed method is compared with the optimum-amplitude method and indirect space vector modulation (SVM), in terms of the input reactive power under different operating conditions, which shows that the proposed method can obtain the maximum input reactive power over most situations.For the two-stage matrix converter (TSMC) topology, the DC-link voltage should always be positive, which leads to a limited control range of input reactive power. Thus, a space vector-inversion method is proposed to realize SVM, which uses opposite vector to replace the original vector. This method can compensate the reactive power limitations to an extent. To further improve the reactive power control properties, a reactive power-extended modulation strategies based on space vector synthesis is presented for the two-stage matrix converter. The operating principle of this method is:first, the output voltage formation and input reactive current formation are completed separately; then, they are combined to form new output reference vectors, with the help of the space vector-inversion method; finally, the combined SVM is used to obtain the desired output voltage and input current. Compared with the existing modulation strategy based on pulse merging, the proposed method has lower computational effort and better reactive power control capability over most situations.Consider the flexible DC-link of the two-stage matrix converter, a novel topology with static synchronous compensator (STATCOM) function is proposed to enhance the input reactive power properties. First, the topology is introduced as well as its working principle, and the related modulation strategy is presented based on the indirect SVM. With this scheme, the input reactive power dependence on the output current can be removed, which would overcome the topological drawback of the limited reactive power control range.As for the matrix converter-based wind energy conversion system (WECS) with doubly-fed induction generator (DFIG), the low voltage ride through (LVRT) issue is also related to the reactive power control. For this issue, a novel scheme based on topology reconfiguration of TSMC is proposed, which adds an auxiliary switching circuit in the DC-link of TSMC. Under the normal network, the auxiliary switching circuit does not work; during the grid faults, the proposed topology would be reconfigured into a current source STATCOM and a general voltage source inverter. With this scheme, the WECS could stay connected to the grid, and provide reactive power to support the grid voltage.更多还原