An Improved Integrated Cumulant Method by Probability Distribution Pre-Identification in Power System With Wind Generation

Bingyan Sui,Jie Xing,Ruilin Cao,Hongyan Ma

doi:10.1109/access.2021.3100627

Bingyan Sui, Jie Xing + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3100627

Copy DOI

Abstract

The increase of wind generation (WG) has challenged the conventional way of probabilistic load flow (PLF) calculation. A reliable and efficient PLF method is required to face the stochastic nature of various power systems with WG. Firstly, the paper analyzes several typical cumulant methods (CMs) for PLF, such as Gram-Charlier expansion of type A (GCA), Gram-Charlier expansion of type C (GCC), and maximum entropy (ME). Then, an improved integrated CM by probability distribution pre-identification is proposed for power systems with WG based on doubly fed induction generations (DFIGs). The skewness and kurtosis are used as probability distribution pre-identification indices in the CM framework. Meanwhile, the influence of the DFIG control strategy on reactive power is considered in the load flow model and the moment calculation. Finally, the accuracy and efficiency of the proposed method are validated with the IEEE test system. In various scenarios, suitable CM is selected and applied to the PLF based on pre-identifying distribution characteristics. Results reveal that probabilistic density functions (PDFs) of bus voltages and line flows obtained by the proposed method have both accuracy and efficiency.

Highlights

In recent years, the development of renewable energy is growing rapidly worldwide, which plays an important role in alleviating fossil energy depletion and environmental pollution [1], [2]
It can be seen that the accuracy of Gram-Charlier expansion of type A (GCA) results is only slightly lower than other cumulant methods (CMs) in some cases and the accuracy for reactive power distributions in line 2-3 and line
PROBABILISTIC RESULTS OF THE STATE VARIABLE DISTRIBUTIONS

Summary

INTRODUCTION

The development of renewable energy is growing rapidly worldwide, which plays an important role in alleviating fossil energy depletion and environmental pollution [1], [2]. R. CAO et al: An improved integrated CM by probability distribution pre-identification in power system with WG state variables of power systems such as bus voltages and line flows meet quasi-normal distributions. CAO et al: An improved integrated CM by probability distribution pre-identification in power system with WG state variables of power systems such as bus voltages and line flows meet quasi-normal distributions In this situation, Gram-Charlier expansion of type A (GCA) is used in the PLF method to fit the probability density function (PDF) of the state variable in power systems with enough accuracy and efficiency [17], [18].

TYPICAL CUMULANT METHODS IN PROBABILISTIC LOAD FLOW

CUMULANTS AND PDFS OF WIND GENERATION BASED ON DFIGS

PRE-IDENTIFICATION INDICES FOR CM SELECTION

CASE STUDY

PROBABILISTIC RESULTS OF THE STATE VARIABLE DISTRIBUTIONS

CONCLUSION