A Strategy for Enhancing the Distributed Photovoltaic Hosting Capacity of Distribution Networks Based on Multiple Management Measures
Year: 2025
Type: article
Abstract: [Objective] Considering the uncertainty of new energy output and user demand in the distribution network， a distributed photovoltaic （PV） hosting-capacity enhancement strategy based on multiple management measures is proposed to enhance the consumption capacity of the distribution network for distributed PVs. [Methods] First， based on active management measures for power and voltage by various regulating devices and demand-side management measures for load reduction， a distributed PV hosting-capacity improvement model considering the stability and economy of distribution-network operation is established. Subsequently， considering the uncertainty of the distributed PV output and user demand， the Wasserstein generative adversarial network （WGAN） scene-generation and K-means scenario-clustering methods are adopted. Through the K-means scenario-clustering method， an uncertain set of probability distributions of source-load characteristic scenarios based on the Kullback-Leibler divergence is constructed. The optimization model for improving the hosting capacity of distributed PV systems is transformed into a distributionally robust optimization model driven by the training data of a deep-learning model. Finally， the effectiveness of the distributed PV hosting-capacity improvement strategy based on multiple management measures for the distributed PV hosting capacity of the distribution network is analyzed via numerical simulation. [Results] The results indicate that the proposed distributed PV hosting-capacity enhancement strategy based on multiple management measures increases the hosting capacity of distributed PV by 2.92% and effectively addresses the randomness of the distributed PV output and load. The autocorrelation and partial correlation coefficients of the distributed PV output and load scenarios generated by the WGAN are highly consistent with historical sample data， thus effectively reflecting their temporal variation characteristics and addressing the difficulty in acquiring historical data. [Conclusion] The proposed strategy not only effectively enhances the hosting capacity of distributed PVs in the distribution network but also provides an effective technical approach to address the uncertainties of renewable-energy outputs and load demands， thus demonstrating its significant theoretical and practical-application significance.
Author: ZHU Chuanpu, CHEN Guanliang, SONG Lizhu, WU Huorong, ZHANG Juncheng, LIAO Xiaobing
Institution: Wuhan Institute of Technology
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7084054799
FWCI:
Cites:
Cited by:
Related to: