电网电压较大跌落下双馈风电系统LVRT控制研究

发布时间：2018-07-05 03:05

本文选题：风力发电 + 电压跌落　；参考：《中国石油大学(华东)》2014年硕士论文

【摘要】：多数大规模风力发电场处于比较偏僻的区域,电网相对薄弱,特别是电网电压存在较大跌落的可能,大规模的风电机组解列将会严重威胁电力系统的安全运行,为此,通常需要风电机组具备电网故障条件下的穿越能力。针对电网故障条件下双馈感应发电机低电压穿越控制的研究已成为近年来的研究热点。双馈感应发电机的控制主要采用矢量控制,获得准确的定子磁链角是矢量控制中实现双馈感应发电机系统准确解耦的关键,理想电网电压条件下,通常采用基于电压模型的磁链观测算法观测定子磁链,然而,传统磁链观测算法大都采用低通滤波器代替纯积分,会引入幅值相位误差,通常需要准确地获取电机同步角频率用于补偿误差或者用于动态调整低通滤波器的截止频率来消除误差。为了提高传统定子磁链观测算法的精度,本文首先研究了双馈感应电机同步角频率估计方法。电网较大跌落条件下,双馈感应电机定子磁链可能会产生负序和暂态直流分量,这将会影响传统定子磁链观测算法的精度和速度,使双馈感应发电机系统无法获得准确解耦,近而影响低电压穿越控制算法的有效实现。为了实现电网故障条件下双馈感应发电机系统的准确解耦,本文进一步研究了新型的定子磁链观测算法。电网较大跌落条件下,定子磁链中的暂态直流分量和负序分量是引起转子过电流的重要原因,为了防止转子侧过电流的出现、确保发电机的安全与稳定运行,必须抑制定子磁链暂态直流分量以及负序分量对发电机转子侧的不良影响。现有的控制策略通常需要提取出定子磁链各序分量进而用于控制,因此,研究电网较大跌落条件下定子磁链各序分量的准确、快速提取算法具有重要意义。本文具体研究内容如下:提出使用二阶广义积分器-锁频环(second order generalized integrator-frequency locked loop,SOGI-FLL)估计异步电机同步角频率的方法,并对其频率响应性能进行理论分析和仿真验证。构建了基于异步电机矢量控制的转速电流双闭环调速系统实验平台,将SOGI-FLL估计出的同步角频率用于磁链观测算法中,实验结果验证了所提方案的有效性。提出使用双二阶广义积分器锁频环(dual second order generalized integratorfrequency locked loop,DSOGI-FLL)估计异步电机同步角频率的新方法,理论分析和实验研究表明,该方法可使同步角频率估计速度不受反电势信号幅值、频率变化的影响,并能消除同步角频率估计结果中2倍频波动成分。提出直接使用DSOGI-FLL进行定子磁链观测的方案,理论分析表明,该方案无需补偿环节,本质为闭环系统且可以有效抑制定子反电势中的高次谐波和低频干扰成分。对该算法的频率响应特性进行了研究,仿真和实验研究表明,该算法具有优异的稳态、动态性能,可实现定子磁链的准确观测。提出电网故障条件下基于DSOGI-FLL的定子磁链正序、负序分量提取算法,理论分析了该算法实现定子磁链各序分量提取的原理,搭建了双馈感应发电机风电系统模型,仿真验证了该算法的有效性。
[Abstract]:Most large-scale wind farms are in a relatively remote area, the power grid is relatively weak, especially the potential of a large drop in the voltage of the power grid. The large-scale wind turbine disassembly will seriously threaten the safe operation of the power system. For this reason, the wind turbine usually needs the crossing ability under the condition of the power grid failure. The research on low voltage crossing control of doubly fed induction generator has become a research hotspot in recent years. The control of doubly fed induction generator is mainly used in vector control. Obtaining accurate stator flux angle is the key to realize accurate decoupling of doubly fed induction generator system in vector control. Under ideal voltage conditions, electricity is usually based on electricity. The magnetic chain observation algorithm of the pressure model observes the stator flux. However, the traditional flux observation algorithms mostly use the low pass filter instead of the pure integral, which will introduce the amplitude phase error. It usually needs to accurately obtain the synchronous angle frequency of the motor to compensate the error or to adjust the cut-off frequency of the low pass filter to eliminate the error. The precision of the traditional stator flux observation algorithm is first studied in this paper. In this paper, the synchronous angular frequency estimation method of doubly fed induction motor is first studied. The stator flux of doubly fed induction motor may produce negative and transient DC components under the condition of large fall, which will affect the precision and speed of the traditional stator flux observation algorithm and make the doubly fed induction generator system In order to realize the accurate decoupling of the doubly fed induction generator system under the condition of the power grid fault, the new stator flux observation algorithm is further studied in this paper. The transient DC component and the negative sequence component in the stator flux linkage are the transient DC component and the negative sequence component. In order to prevent the overcurrent of the rotor, in order to prevent the occurrence of the rotor side overcurrent and ensure the safe and stable operation of the generator, the transient DC component of the stator flux and the negative effect of the negative sequence component on the rotor side of the generator must be suppressed. The current control strategy usually needs to extract the sequence components of the stator flux and then use it for control. Therefore, it is of great significance to study the accuracy of the sequence components of the stator flux under the condition of large drop in the power grid. The contents of this paper are as follows: the method of using the Nikai Hiroyoshi integrator - second order generalized integrator-frequency locked loop, SOGI-FLL to estimate the synchronous angle frequency of the asynchronous motor is proposed. The experimental platform of speed and current double closed loop speed control system based on vector control of asynchronous motor is constructed. The synchronous angular frequency estimated by SOGI-FLL is used in the magnetic chain observation algorithm. The experimental results verify the effectiveness of the proposed scheme. A double two order generalized integrator lock is proposed. The frequency ring (dual second order generalized integratorfrequency locked loop, DSOGI-FLL) is a new method for estimating synchronous angular frequency of asynchronous motor. The theoretical analysis and experimental study show that the method can make the speed of synchronous angular frequency estimation not affected by the amplitude of the back EMF signal and the frequency change, and can eliminate the 2 times of the frequency estimation results of the synchronous angle. The frequency fluctuation component. A scheme of stator flux observation using DSOGI-FLL is proposed. The theoretical analysis shows that the scheme does not need compensation link. It is essentially closed loop system and can effectively suppress the high harmonic and low frequency interference components in the stator back EMF. The frequency response characteristics of the algorithm are studied. Simulation and experimental research show that The algorithm has excellent steady state and dynamic performance, and can realize the accurate observation of stator flux. The algorithm of positive sequence and negative sequence component extraction based on DSOGI-FLL based stator flux is proposed under the condition of power grid fault. The principle of realizing the extraction of each sequence component of stator flux is theoretically analyzed, and the model of the doubly fed induction generator wind power system is built, and the simulation is verified. The effectiveness of the algorithm is proved.
【学位授予单位】：中国石油大学(华东)
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM346

【相似文献】