Research on classification and determination of medium-to-long-term flexibility capacity requirements in high-proportion renewable energy power systems based on seasonal scales
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Abstract With the rapid growth of new energy installed capacity in China, the fluctuating and uncertain output of wind and solar power has intensified the regulatory pressure on power systems. Traditional capacity planning methods based on annual average load struggle to accurately reflect seasonal variation characteristics. To address this, this paper proposes a seasonal-scale classification method for medium-to-long-term flexibility capacity requirements in power systems with high new energy penetration. First, the K-means clustering algorithm is employed to divide annual operational states into two typical intervals: summer and non-summer, capturing source-load seasonal differences. Second, a random forest model predicts renewable generation and load, calculates net load, and analyzes its three-hour ramping characteristics. This establishes a three-tier classification system: basic flexibility (C 1 ), peak flexibility (C 2 ), and ultra-peak flexibility (C 3 ). Finally, the methodology’s validity was validated using the California power system as a case study, followed by application to three Chinese provinces: Jiangsu, Ningxia, and Xinjiang. These findings provide quantitative evidence and decision-making references for medium-to-long-term capacity planning and prioritized allocation of flexibility resources in power systems with high renewable energy penetration.
Fields of Study & Subjects
Electric Power System Optimization
Integrated Energy Systems Optimization
Power Systems and Renewable Energy
Publication details
- Journal / source
- Journal of Physics Conference Series
- Volume
- 3198
- Issue
- 1
- Pages
- 012007–012007
- Publication date
- 1 March 2026
- Language
- EN