Improvement of Probabilistic Load Flow to Consider Network Configuration Uncertainties

Abstract

An improved probabilistic load flow method considering network configuration uncertainties is proposed in this paper. The linear relationship between power injections and line flows is deduced, which enables applying cumulants and Gram-Charlier expansion to compute the probabilistic distributions of the transmission line flows, instead of complex convolution calculations. Then compensation method and total probability theorem are combined to model the random variations of network configuration. In this way, the probabilistic distributions of state variables can be easily obtained, considering the random factors of loads, generators and network configuration. The case study of WSCC 179-node test system shows that the uncertainties of network configuration remarkably affect the probabilistic distributions of state variables. Thus the proposed method could provide planners with more accurate and comprehensive information. Furthermore, the comparison with Monte Carlo simulation verifies its higher speed, high degree of accuracy and the prospects for practical application.