Global Sensitivity Analysis for Power System with Correlated Inputs Based on the Modified ANCOVA Indices

The existence of a large number of uncertainty sources such as renewable energy plants makes the probability analysis of power system becomes extremely important. As a step of uncertainty analysis, the Global Sensitivity Analysis (GSA) has been applied to power system in recent years. However, the existing analysis of covariance (ANCOVA) method which aims to generate a kind of sensitivity indices for correlated random input variables using the covariance decomposition method is unclearly defined and excludes the total sensitivity indices. Therefore, in this paper, the modified ANCOVA-based indices are introduced to reach the implementation of probabilistic power flow (PPF) for the first time. Then, the basis-adaptive sparse polynomial chaos expansion (BASPCE) is chosen to efficiently calculate the above indices. The efficiency of GSA on PPF by BASPCE method, with the modified indices, is verified on the IEEE 14-bus system and the IEEE 118-bus system with wind farms in the case of correlated inputs. And its analysis results are compared with those of the commonly used Monte Carlo simulations (MCS) method.