摘要
准确的风电功率超短期预测对大规模风电并网运行具有重要意义,该文根据预测模式将风电功率超短期预测分为滚动直接预测、滚动多步预测和多采样间隔滚动预测3种。在初步分析的基础上对各预测模式进行针对于混沌序列初值敏感性的适用性改进,使序列的初值相对恒定,从而提高其预测精度。最后,利用吉林省某风电场群的实测数据对改进前后的各混沌预测模型进行分析。算例结果表明,改进后的多采样间隔混沌预测模式具有较好的预测效果,且相较改进前的预测效果有大幅提升,从而验证了所提预测模式改进方法对于提升超短期混沌预测精度的有效性。
Accurate wind power ultra-short-term prediction is important for large-scale wind power grid-connected operation. This paper classified wind power ultra-short-term prediction into three types according to prediction modes:rolling direct prediction, rolling multi-step prediction and multi-sampling interval rolling prediction. Based on the preliminary analysis, the applicability of each prediction mode was improved for the initial value sensitivity of chaotic sequences, so that the initial values of the sequences were relatively constant, thus improving their prediction accuracy.Finally, each chaotic prediction model was analyzed before and after the improvement using the measured data of a wind farm cluster in Jilin Province, China. The results show that the improved multi-sampling interval chaotic prediction model has a better prediction effect, and the prediction effect is significantly improved compared with the prediction effect before the improvement, thus verifying the effectiveness of the improved prediction model to improve the accuracy of ultra-short-term chaotic prediction.
作者
杨茂
孙志博
苏欣
YANG Mao;SUN Zhibo;SU Xin(Key Laboratory of Modern Power System Simulation and Control&Renewable Energy Technology,Ministry of Education(Northeast Electric Power University),Jilin 132012,Jilin Province,China)
出处
《中国电机工程学报》
EI
CSCD
北大核心
2022年第22期8117-8128,共12页
Proceedings of the CSEE
基金
国家重点研发计划项目(2018YFB0904200)。
关键词
风电功率
超短期预测
混沌理论
预测模式
多采样间隔
wind power
ultra-short-term forecast
chaos theory
prediction model
multi-sampling interval
