Open Access

ARTICLE

Distributionally Robust Co-Optimization of Rural PV-Storage Distribution Networks Incorporating Adjustable Agricultural Load Characteristics

Xiaole Luo^1,*, Haitao Lan², Yang Song¹, Yunliang Zhao³, Yingnan Bai³, Yinan Yang⁴, Haichao Peng⁵, Xiaojing Wang^6,7
1 Marketing Department, State Grid Songyuan Power Supply Company, Songyuan, China
2 Marketing Department, State Grid Jilin Electric Power Co., Ltd., Changchun, China
3 Qianguo County Power Supply Center, State Grid Songyuan Power Supply Company, Songyuan, China
4 Marketing Service Center, State Grid Jilin Electric Power Co., Ltd., Changchun, China
5 Science, Technology and Internet Department, State Grid Songyuan Power Supply Company, Songyuan, China
6 College of Electrical Engineering, Zhejiang University, Hangzhou, China
7 Zhejiang Zheda Energy Science & Technology Co., Ltd., Hangzhou, China
* Corresponding Author: Xiaole Luo. Email:
(This article belongs to the Special Issue: Advances in Energy Modelling for Sustainable, Intelligent and Resilient Power and Energy Systems)

Energy Engineering https://doi.org/10.32604/ee.2026.079356

Received 20 January 2026; Accepted 09 March 2026; Published online 19 May 2026

Abstract

Rural distribution networks are experiencing increasing integration of distributed photovoltaic (PV) generation, yet the inherent intermittency of PV output and the pronounced peak-valley disparities of agricultural loads pose significant challenges to grid economy and security. To address photovoltaic (PV) fluctuations and agricultural load peak-valley disparities in rural distribution networks, this paper proposes a distributionally robust cooperative planning method integrating load elasticity. Firstly, a refined agricultural load model is established: irrigation loads couple pump head-flow characteristics with load shifting mechanisms; greenhouse loads incorporate curtailable heating and temperature-light constraints; production loads combine time-shiftable machinery charging with rigid sensor demands. A two-stage distributionally robust optimization model is then formulated to minimize total investment-operation costs under PV uncertainty, innovatively embedding the CMA-ES algorithm for coordinated electricity consumption strategy optimization. Simulations on a modified IEEE 33-node system demonstrate: compared to conventional planning, the proposed method reduces total costs by 15.6% and mitigates peak-valley differences by 28.5% while maintaining voltage deviations within 1.04–1.05 p.u., verifying the synergy of distributed PV-storage systems and load elasticity.

---

Keywords

Load elasticity; distributed PV-storage; distributionally robust optimization; rural distribution network; CMA-ES algorithm

---