Day-ahead and intra-day coordinated optimal scheduling for photovoltaic active distribution networks based on SARIMA prediction model

Authors

Yang ZHAO, Economic and Technological Research Institute, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, ChinaFollow
Zhifang HAO, Economic and Technological Research Institute, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, China
Cong WANG, Shiji azhuang Power Supply Branch, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, China
Xiaolin TAN, Economic and Technological Research Institute, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, China
Zihao ZHAO, Economic and Technological Research Institute, State Grid Hebei Electric Power Co.,Ltd., Shijiazhuang 050081, China
Wei GUO, Economic and Technological Research Institute, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, China
Wenyi FAN, Economic and Technological Research Institute, State Grid Hebei Electric Power Co., Ltd., Shijiazhuang 050081, China

Keywords

distribution network scheduling; SARIMA prediction; affine optimization; adjustable robust; modelpredictive control

Abstract

A day-ahead and intra-day two-stage op timal scheduling method for active distribution networks based on the seasonal autoregressive integral moving average （SARIMA ） prediction model is proposed to solve the problem that the scheduling of traditional photovoltaic （PV） active distribution networks cannot fully mine the historical data of distributed PV output.Meanwhile, the affine relationship between distributed PV output prediction and distribution network optimal scheduling is established.Firstly, the historical data of distributed PV output are adopted to construct the interval boundary for PV output based on SARIMA prediction confidence level, thus forming the optimization interval for the uncertainty optimization problem.Then, by establishing the affine mechanism between confidence level and decision variables, a day-ahead and intra-day two-stage optimal scheduling model for distribution networks is built, where the optimization problem is transformed into a mixed integer programming problem for solution via the linear relaxation of affine variables.Finally, the effectiveness of the proposed method is verified by the improved IEEE 33-node distribution system, with the operation costs of the system at different confidence levels compared.

DOI

10.19781/j.issn.1673-9140.2025.05.005

First Page

46

Last Page

58

Recommended Citation

ZHAO, Yang; HAO, Zhifang; WANG, Cong; TAN, Xiaolin; ZHAO, Zihao; GUO, Wei; and FAN, Wenyi (2025) "Day-ahead and intra-day coordinated optimal scheduling for photovoltaic active distribution networks based on SARIMA prediction model," Journal of Electric Power Science and Technology: Vol. 40: Iss. 5, Article 5.
DOI: 10.19781/j.issn.1673-9140.2025.05.005
Available at: https://jepst.researchcommons.org/journal/vol40/iss5/5