@Article{ee.2026.088744,
AUTHOR = {Bin Lin, Jingmiao Huang, Quanmin Chen, Sihui Ke, Yusi Xu, Yan Zhang},
TITLE = {Dirichlet Process Mixture Model-Based Affine Voltage-Risk-Aware Planning of Distributed Photovoltaic Generation under Source-Load Uncertainty},
JOURNAL = {Energy Engineering},
VOLUME = {},
YEAR = {},
NUMBER = {},
PAGES = {{pages}},
URL = {http://www.techscience.com/energy/online/detail/28015},
ISSN = {1546-0118},
ABSTRACT = {High penetration of distributed photovoltaic (PV) generation increases the impact of source-load uncertainty on voltage security and economic performance in distribution network planning. Planning methods that neglect uncertainty may underestimate voltage risks caused by PV and load fluctuations and may not fully reflect the performance differences among candidate PV schemes under different operating scenarios. To address this issue, this paper develops a distributed PV planning method that combines Dirichlet process mixture model (DPMM)-based scenario modeling with affine voltage risk evaluation. Historical PV and load data are organized into daily joint PV-load samples, and DPMM is used to identify latent source-load operating components. Based on these components, representative scenarios are constructed, from which occurrence probabilities, central profiles, and quantile-based fluctuation bounds are obtained. The scenario centers and fluctuation bounds are then converted into affine power injections, so that intra-scenario fluctuations can be propagated to nodal voltage intervals through affine power flow. In the proposed planning and evaluation framework, candidate PV siting and sizing schemes are assessed using scenario-weighted economic cost and affine voltage-risk indicators, and particle swarm optimization is used to update the planning decisions. Case studies on the IEEE 69-bus system and a practical 80-bus distribution network, together with annual Monte Carlo validation, indicate that the proposed method can incorporate uncertainty-induced voltage risk and multi-scenario economic performance into PV planning, thereby supporting a balanced trade-off between economic performance and voltage security.},
DOI = {10.32604/ee.2026.088744}
}