首页|期刊导航|太阳能|中国风能和太阳能发电量增长的稳定域和波动域划分与预测研究

中国风能和太阳能发电量增长的稳定域和波动域划分与预测研究OA

STUDY ON DELINEATION AND FORECASTING FOR STABLE DOMAIN AND FLUCTUATING DOMAIN OF WIND AND SOLAR POWER GENERATION CAPACITY GROWTH IN CHINA

中文摘要英文摘要

鉴于风能和太阳能发电量增长的稳定部分和季度性波动部分在电力系统规划定位、风险管理和驱动因素上均呈现出异质性,提出划分"稳定域+波动域"的新方法.针对中国风能和太阳能发电量统计数据的"小样本、快变化"特征,采用新陈代谢 GM(1,1)模型作为核心工具,构建了稳定域与波动域相结合的预测方法,并基于 2017-2024 年中国风能和太阳能发电量季度数据进行了模型预测精度测试.研究结果表明:1)训练阶段和测试阶段风能和太阳能发电量预测的平均绝对百分比误差(MAPE)均低于 5%,验证了本文所提方法的有效性.2)根据预测结果,"十五五"期间,中国风能和太阳能发电量的稳定域将保持指数型增长,其中,太阳能发电量和风能发电量稳定域的年复合增长率将分别达到 26.98%和 23.30%;与2024 年相比,2030 年 4 个季度的风能发电量的波动值均增长 2 倍以上,其中,第 4 季度增长近 2.5 倍;太阳能发电量的波动值增幅超过 3 倍."稳定域+波动域"划分方式及预测方法,为新型能源体系中长期规划与风险评估提供了理论参考.

Given the heterogeneity in power system planning orientation,risk management,and driving factors between the stable component and the seasonal fluctuation component of wind and solar power generation capacity growth,this paper proposes a novel"stable domain+fluctuating domain"partitioning paradigm.To address the"small sample size,fast-changing"characteristics of China's wind and solar power generation capacity statistics,the metabolism GM(1,1)model is adopted as the core tool to construct a prediction method for the stable and fluctuating domain.An empirical study based on national quarterly data of wind and solar power generation capacity from 2017 to 2024.The research results show that:1)The mean absolute percentage error(MAPE)of wind and solar power generation capacity prediction is below 5% on both the training set and the test set,verifying the method's effectiveness.2)The prediction results indicate that during the 15th Five-Year plan period,the stable domain of China's wind and solar power generation capacity will maintain exponential growth,with the compound annual growth rates(CAGR)of the stable domain for wind and solar power generaation capacity reaching approximately 27% and 23%,respectively.Compared with 2024,the fluctuation value of wind power generation capacity in the four quarters of 2030 has increased by more than 2 times,with a nearly 2.5 times increase in the Q4.The quarterly fluctuation of solar power generation capacity has increased by about 3 times.The partitioning paradigm and prediction method proposed in this study provide a theoretical basis and reference for medium-to long-term planning and risk assessment in the new energy system.

王洁方;魏佳丽

华北水利水电大学管理与经济学院,郑州 450046华北水利水电大学管理与经济学院,郑州 450046

信息技术与安全科学

风能太阳能稳定域波动域预测灰色建模

wind energysolar energystable domainfluctuating domainforecastinggrey modeling

《太阳能》 2026 (9)

24-34,11

国家社科基金一般项目(22BJY134)

10.19911/j.1003-0417.tyn20251206.01

评论