Source–Load Data-Driven Multi-Objective Siting and Sizing of Distributed Energy Storage in Islanded Microgrids
The variability of renewable generation and load demand complicates energy storage planning for islanded microgrids, particularly when siting, sizing, operating reliability, and economic performance must be considered simultaneously. This paper proposes a source-load data-driven method for siting and sizing distributed energy storage in an islanded microgrid. An improved grey wolf optimiser-gated recurrent unit (IGWO-GRU) model was developed to forecast photovoltaic output and nodal loads under normal and adverse weather conditions. The forecasts were incorporated into a multi-objective planning model that maximised islanded supply reliability and capacity-normalised discharge throughput while minimising annualised planning and operating cost. Continuous state-of-charge evolution, transitions between grid-connected and islanded operation, nodal-voltage limits, line-loading limits, upstream-grid power exchange, and priority supply to critical loads were explicitly considered. NSGA-III was directly coupled with a pandapower full-chronology AC power-flow model so that every candidate configuration was evaluated from its simulated operating performance. In the synthetic PV–BESS IEEE 33-bus case study, the 1319 h chronological assessment yielded 640 feasible configurations, including 10 non-dominated solutions. For this synthetic case and predefined candidate buses, the results illustrate a chronological planning workflow and quantify conditional trade-offs among annualised cost, islanded supply reliability, and capacity utilisation. They do not establish universal forecasting or optimisation superiority.
Authors
- Guanglin Sha (ORCID: https://orcid.org/0000-0002-1801-115X)
- Tao Zhou (ORCID: https://orcid.org/0000-0003-0561-2316)
- Zhong Wang (ORCID: https://orcid.org/0000-0002-6527-6601)
- Yang Yi (ORCID: https://orcid.org/0009-0009-0562-2363)
- Haijun Yu (ORCID: https://orcid.org/0009-0000-0519-1350)
- Jianfang Chen
- Jianming Su
- Guansen Wang
- Guangjing Zhou
- Yunzhao Wu
Institutions
- State Grid Corporation of China (China) (CN)
- Beijing Jiaotong University (CN)
- Lanzhou Jiaotong University (CN)
- Shanghai Electric (China) (CN)
- China Electric Power Research Institute (CN)
Publication Details
- Journal
- Applied Sciences
- Published
- 2026-11-18
- DOI
- https://doi.org/10.3390/app162211448
- Primary Topic
- Optimal Power Flow Distribution
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Science and Technology Project of State Grid