基于谐波最大似然估计的分布式光伏接入电网承载能力评估新方法

Methods for fully characterizing the harmonic injection amount of distributed photovoltaic grid connection and combining harmonic constraints with other constraints to accurately evaluate the hosting capacity of photovoltaic integration into distribution networks are of great significance as they ensure the safe and stable operation of distribution networks. Therefore, a novel hosting capacity evaluation method for distributed photovoltaics (PVs) connected in a power system based on the maximum likelihood estimation of harmonics (MLE) is proposed in this study. First, using the likelihood function from the MLE method, the harmonic parameters of distributed photovoltaic injections are optimally estimated, allowing for the accurate assessment of harmonic outputs during photovoltaic grid connections. Furthermore, a harmonic partitioning method is devised; it characterizes the connection degree between nodes in the grid-connected system, and it divides the distribution network into regions. The scenario number in the estimation of hosting capacities is effectively reduced. Finally, a comparison is carried out relative to the conventional hosting capacity. The assessment method proposed in this study considers the harmonic access in the actual distributed PV grid-connected system. An improved harmonic partitioning method is established based on the harmonic injection amount. The evaluation of PV hosting capacities in the region ensures accuracy and reduces calculation times. They provide references for the access capacity of the distribution network.