电力系统及其自动化学报2024，Vol.36Issue(8)：133-141,9.DOI:10.19635/j.cnki.csu-epsa.001430
结合数据增强及组合算法的短期光伏功率预测
Short-term Photovoltaic Power Prediction Using Data Augmentation and Combined Algorithms
摘要
Abstract
A short-term photovoltaic(PV)power prediction model based on data augmentation and combined algorithmsis proposed to address the problems of insufficient data completeness and low power prediction accuracy in PV power generation data. First,the PV data is divided into different weather types using the K-means++clustering algorithm. Second,conditional generative adversarial networks are used to learn the distribution patterns of PV data and generate high-quality samples. Third,the decomposition number and penalty factor of variational mode decomposition are opti-mized to further reduce the fuzzy entropy value of subsequences. Finally,the input weights and biases of the deep ex-treme learning machine are optimized by using the sine cosine algorithm,and predictive models are established for each subsequence. Experimental results show that the proposed model has its superiority关键词
光伏功率预测/数据增强/变分模态分解/深度极限学习机/正余弦算法Key words
photovoltaic(PV) power prediction/data augmentation/variational mode decomposition/deep extremelearning machine/sine cosine algorithm分类
信息技术与安全科学引用本文复制引用
毛嘉铭,刘光宇,罗凯元..结合数据增强及组合算法的短期光伏功率预测[J].电力系统及其自动化学报,2024,36(8):133-141,9.基金项目
国家自然科学基金资助项目(62273124) （62273124）
杭州电子科技大学学生科学研究基金资助项目(CXJJ2023202) （CXJJ2023202）
