随着大量新能源场站接入交直流混联电网,系统的静态电压稳定裕度(static voltage stability margin,SVSM)水平具有很大的不确定性,需要研究考虑新能源场站高阶不确定性的交直流混联电网SVSM计算方法.针对此问题,首先建立了交直流混联电网SVSM计算模型,模型中考虑了直流换流站控制方式随负荷增长的切换;采用概率盒模型描述风速与光照强度的随机波动,提出了改进区间半不变量法以获得更准确的SVSM概率盒,该方法通过K-means++聚类算法将随机变量样本划分为多个波动范围较小的样本集,以降低半不变量的线性化计算带来的误差;并结合Gram-Charlier级数展开和概率加权和计算得到考虑新能源场站高阶不确定性的系统SVSM概率盒.通过对修改的IEEE-39节点交直流系统和南方电网两个算例的分析,并与区间半不变量法和双层蒙特卡洛法比较,验证了所提出方法获得的SVSM概率盒具有较高的计算精度和效率.
Calculation of Static Voltage Stability Margin for AC/DC Hybrid Power System Considering the High-order Uncertainty of Renewable Energy
As lots of renewable energy stations are integrated into AC/DC hybrid power grid, the static voltage stability margin of the system has greater uncertainty. It is necessary to study the static voltage stability margin(SVSM) calculation method of the AC/DC hybrid power grid considering the high-order uncertainty of renewable energy. To address this problem, an SVSM calculation model for AC/DC hybrid power grid is established, which considers the control mode switching of DC converter stations with the load growth. A probability box model is used to describe the random fluctua-tions of wind speed and light intensity, and an improved interval cumulant method is proposed to obtain a more accurate SVSM probability box. This method uses the K-means++ clustering algorithm to divide random variable samples into multiple sample sets with smaller fluctuation ranges to reduce the errors due to the linearization calculation of cumulants. Combined with the Gram-Charlier series expansion and the probability weighted sum calculation, the SVSM probability box considering the high-order uncertainty of renewable energy is obtained. Case studies on the modified IEEE-39 bus AC/DC system and China Southern Power Grid, comparedwith interval cumulant method and two-level Monte Carlo method, demonstrate that the SVSM probability box obtained by the proposed method has high calculation accuracy and efficiency.
renewable energyAC/DC hybrid power gridhigh order uncertaintystatic voltage stability marginparameterized probability boximproved interval cumulant method