Bi-Level Sizing Method for Wind Farm Energy Storage Considering Market Benefits of Total Life Cycle

The combined output of wind power and energy storage can correctly enhance the profits of the wind storage system. However, the capacity of the energy storage system (ESS) and the joint output method of the wind farm with ESS immediately affect the economy of the joint device of the wind farm with E...

Full description

Saved in:
Bibliographic Details
Published in2022 4th International Conference on Electrical Engineering and Control Technologies (CEECT) pp. 907 - 913
Main Authors Cai, Zhongqi, Yang, Kun, Gu, Yanxun, Lin, Guihui, Liu, Chao, Sun, Sashuang
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2022
Subjects
Online AccessGet full text
DOI10.1109/CEECT55960.2022.10030682

Cover

More Information
Summary:The combined output of wind power and energy storage can correctly enhance the profits of the wind storage system. However, the capacity of the energy storage system (ESS) and the joint output method of the wind farm with ESS immediately affect the economy of the joint device of the wind farm with ESS to participate within the marketplace. Therefore, this paper proposes a bilevel sizing method for wind farm energy storage considering the full life cycle market operation benefits. The approach is based on a two-layer optimization model. The outer layer takes the maximum revenue from the joint system's participation within the day-in-advance marketplace and the minimal life-cycle cost of ESS as the objective function, and the inner layer takes the maximum operating revenue of the 8760h day-ahead market as the objective function, taking into consideration the effect of the life loss of electricity storage. Then the improved particle swarm optimization algorithm and CPLEX are used to collectively remedy the problem. This approach utterly considers the coupling relationship between ESS sizing and market operation and additionally considers the genuine loss of ESS in the planning, which is intently associated with reality. Ultimately, the numerical effects affirm the effectiveness of the proposed method.
DOI:10.1109/CEECT55960.2022.10030682