An Improved Cumulant Method for Probabilistic Load Flow Calculation

This paper proposes an improved cumulant method for probabilistic load flow calculation. The proposed method combined cumulant method and K-means clustering technique. Firstly, the output power of wind farms is divided into several clusters by K-means technique. Then, the conventional cumulant method is performed in each cluster to obtain the probability distributions of output random variables. Finally, the probabilistic load flow results are obtained by integrating probability distributions of output random variables and corresponding weight of each cluster. The performance of the proposed method is evaluated on a modified IEEE 30-bus system. The simulation results demonstrate that the proposed method could yield accurate results with a low computational burden, and it also overcomes the limitation of existing series expansion methods to approximate multimodal probability distributions.