Reactive Voltage Partitioning Method for the Power Grid With Comprehensive Consideration of Wind Power Fluctuation and Uncertainty

Xu Zhang,Yixian Wang,Xueting Cheng,Yunlong Chen,Xiaobing Zha,Yi Wang,Yuchuan Zheng,Ruiting Ding

doi:10.1109/access.2020.3004484

Abstract

When large-scale wind power is connected to the power grid, the fluctuation and uncertainty in the wind power reduce the stability and accuracy of the grid's reactive voltage division results based on the electrical distance matrix and affect the grid's reactive power regulation. This paper proposes a grid reactive voltage partitioning method that considers the wind power stability and accuracy in a comprehensive manner. The wind power uncertainty and zoning results are characterized by the distribution of wind power forecast error intervals and changes in the zoning result nodes at different moments when the wind power is connected. Regarding volatility, according to the discretization of the probability distribution of the active power output at a certain time based on the wind power prediction, a calculation interval of the wind power output under a single cross-section is formed, and multiple sequential power flow sections within a long time scale are clustered and partitioned by an agglomeration hierarchical clustering method. Finally, an optimal zoning model of reactive voltage is established over a long time scale with the minimum comprehensive stability serving as the objective function. A simulation analysis of the improved IEEE39 node system shows that the partition combination can effectively increase the stability and accuracy of the reactive partitioning.

Highlights

With the increasing scale of grid interconnection and improvements in online automation, automated voltage control (AVC) in grid hierarchical partitioning has been widely applied to ensure the safety and stability of grid operation
There are mainly three modes of reactive power hierarchical voltage control, namely, the three-stage voltage control mode first proposed by EDF, the two-stage voltage control mode proposed by Deutsche Power and the ‘‘soft’’ three-stage voltage control mode that is widely applied in China
After wind power and other new energy sources are connected to the power grid, the current research plan is mainly to replace the electric distance matrix under a single power flow section by equating the volatility of active wind power in the electric distance expectation matrix through a probabilistic method to reduce the influence of wind power fluctuations on the partition stability to a certain extent

Summary

INTRODUCTION

With the increasing scale of grid interconnection and improvements in online automation, automated voltage control (AVC) in grid hierarchical partitioning has been widely applied to ensure the safety and stability of grid operation. After wind power and other new energy sources are connected to the power grid, the current research plan is mainly to replace the electric distance matrix under a single power flow section by equating the volatility of active wind power in the electric distance expectation matrix through a probabilistic method to reduce the influence of wind power fluctuations on the partition stability to a certain extent. This approach is based mainly on the probabilistic statistics of the historical output data of wind farms, so accurate real-time partition results cannot be obtained. The accuracy of power grid dispatching is effectively improved

THE ESTABLISHMENT OF AN ALL-DIMENSIONAL

THE AGGLOMERATIVE HIERARCHICAL CLUSTERING METHOD

2) EVALUATION INDEX OF CLUSTERING ALGORITHM

1) OBJECTIVE FUNCTION

CONCLUSION