双馈并网风力发电机预测直接功率控制的研究

发布时间：2018-06-15 00:32

本文选题：直接功率控制 + 预测功率控制　；参考：《燕山大学》2014年硕士论文

【摘要】：随着社会与经济的高速发展，常规能源日趋紧缺，大力发展风能等可再生能源是可持续能源战略发展的必然趋势。风力发电系统中，双馈感应发电机（DFIG）由于成本低、体积小，不存在永磁材料中的退磁隐患等原因成为风电系统中的主流机型。近年来，直接功率控制（DPC）因为控制简单、解耦性好、动态响应快、鲁棒性强等优点越来越受到关注。本论文对DFIG系统的预测直接功率控制（P-DPC）做了研究，研究的主要内容概括如下：本文将DFIG的控制分为网侧变换器（GSC）控制和转子侧变换器（RSC）控制两个单独的部分。首先，介绍了传统查表法直接功率控制（LUT-DPC），这种方法控制结构简单、动态响应快，但是电流谐波含量大，开关频率不固定。针对这个问题，预测直接功率控制对传统直接功率控制做了改进，本文分别研究了三矢量预测直接功率控制（TV-PDPC）和预测功率优化控制（PPOC），这两种方法都是以有功和无功功率误差最小为目标函数进行控制，TV-PDPC直接计算空间电压矢量的作用时间，而PPOC则计算SVPWM的最优空间矢量来实现控制目标。这两种预测控制都大大减小了电网电流的谐波含量，，实现了开关频率的固定。对于DFIG的并网控制，采用直接虚功率控制（DVPC），该方法利用虚拟的有功无功功率代替并网前恒定为零的定子有功无功功率，形成反馈通道。该方法定子电压与电网电压能很快同步，并网过程平滑，定转子基本无冲击电流，满足软并网的要求。电网电压不平衡时，有功无功功率虽然能保持恒定，但是此时电流谐波含量大，不符合电网相关规定。本文针对不平衡电网条件下的直接功率控制，改进了算法。这种改进方法控制简单，只需要提取负序电压与正序电流，避免了提取负序电流，不需要改变控制结构，可以与电网电压平衡时共用一套控制结构。最后，通过MATLAB/Simulik仿真软件和dSPACE实验平台对以上理论分析以及控制策略进行了仿真和部分实验验证，仿真和实验结果表明所提到的预测直接功率控制具有良好的稳、动态性能，验证了所研究算法的正确性和可行性。
[Abstract]:With the rapid development of society and economy, conventional energy is becoming increasingly scarce. It is an inevitable trend of sustainable energy strategy to develop renewable energy such as wind energy. In wind power system, doubly-fed induction generator (DFIGG) has become the mainstream type of wind power system because of its low cost, small volume and no hidden danger of demagnetization in permanent magnet materials. In recent years, direct power control (DPC) has attracted more and more attention because of its simple control, good decoupling, fast dynamic response and strong robustness. In this paper, the predictive direct power control (P-DPC) of DFIG system is studied. The main contents of the study are summarized as follows: in this paper, the DFIG control is divided into two separate parts: grid-side converter / GSC-control and rotor-side converter / RSC-control. First of all, the traditional direct power control (DPC) based on look-up table method is introduced. The control structure is simple, the dynamic response is fast, but the current harmonic content is large and the switching frequency is not fixed. In order to solve this problem, predictive direct power control improves the traditional direct power control. In this paper, the three vector predictive direct power control (TV-PDPC) and the predictive power optimal control (PPOCPC) are studied respectively. These two methods take the minimum error of active and reactive power as the objective function to control TV-PDPC to calculate the working time of space voltage vector directly. PPOC computes the optimal space vector of SVPWM to achieve the control objective. These two predictive controls can greatly reduce the harmonic content of power network current and realize the fixed switching frequency. For the grid-connected control of DFIG, direct virtual power control is used to control DVPC.virtual active and reactive power is used instead of zero stator active and reactive power to form feedback channel. This method can synchronize the stator voltage with the grid voltage quickly, the process of grid connection is smooth, the stator and rotor have no impulse current, so it can meet the requirement of soft grid connection. When the voltage is unbalanced, the active power and reactive power can be kept constant, but the current harmonic content is large, which does not conform to the relevant regulations of the power network. In this paper, the algorithm is improved for direct power control under unbalanced power system. This improved method is simple to control, only needs to extract negative sequence voltage and positive sequence current, avoids extracting negative sequence current, does not need to change the control structure, and can share a set of control structure with power grid voltage balance. Finally, through MATLAB / Simulik simulation software and DSpace experimental platform, the theoretical analysis and control strategy are simulated and some experimental results are given. The simulation and experimental results show that the proposed predictive direct power control has good stability and dynamic performance. The correctness and feasibility of the proposed algorithm are verified.
【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM315;TM46

【参考文献】