多模型自适应控制应用于大型光伏电站阻尼区间振荡

发布时间：2018-06-16 17:55

本文选题：大型光伏电站 + 区间振荡　；参考：《中国电机工程学报》2017年14期

【摘要】：研究表明对大型光伏电站引入附加阻尼控制,可以有效抑制互联电力系统的区间低频振荡。为解决传统阻尼控制方法在应对大扰动方面的不足,引入多模型自适应控制策略进行阻尼控制。采用K-medoids算法以低频振荡模式为多维特征点对运行状态对应的线性模型进行聚类,并基于各类的中心模型分别设计阻尼控制器。通过比较实际系统与模型的输出动态响应差异,用贝叶斯方法实时计算每一个模型匹配当前实际系统的概率,据此更新各阻尼控制器输出权重,各控制器输出的加权平均作为最终控制信号附加于光伏无功控制的参考指令中。在DIgSILENT/PowerFactory中搭建光伏并网四机两区域测试系统,非线性仿真结果表明,在无需系统受扰后状态先验知识的情况下,该策略可以有效抑制非预期扰动引发的区间振荡,并在反馈信号时延和光照随机波动下均保持良好的鲁棒控制性能。
[Abstract]:It is shown that the application of additional damping control to large PV power plants can effectively suppress the interval low frequency oscillation of interconnected power systems. In order to solve the deficiency of traditional damping control method in dealing with large disturbance, a multi-model adaptive control strategy is introduced to control damping. The K-medoids algorithm is used to cluster the linear model corresponding to the running state with the low frequency oscillation mode as the multidimensional feature point, and the damping controller is designed based on each kind of central model. By comparing the dynamic response difference between the real system and the model, the Bayesian method is used to calculate the probability of each model matching the actual system in real time, according to which the output weights of each damping controller are updated. The weighted average output of each controller is attached to the reference instruction of photovoltaic reactive power control as the final control signal. A two-area photovoltaic grid-connected two-area test system is built in the DIgSILENT / PowerFactory. The nonlinear simulation results show that the strategy can effectively suppress the interval oscillation caused by the unexpected disturbance without prior knowledge of the system state after disturbance. And the robust control performance is maintained under the feedback signal delay and the illumination random fluctuation.
【作者单位】：输配电装备及系统安全与新技术国家重点实验室(重庆大学);
【基金】：国家自然科学基金项目(51477021)~~
【分类号】：TM615;TM712

【参考文献】