基于多尺度熵-动态对齐融合法结合AFGRU的储能调频指令预测
Prediction of Energy Storage Frequency Regulation Commands Based on Multi-scale Entropy Dynamic Alignment Fusion Method Combined with AFGRU
- 2026年47卷第2期 页码:37-44
DOI:10.14044/j.1674-1757.pcrpc.2026.02.004
移动端阅览
西安热工研究院有限公司,西安 710048
华能伊敏煤电有限公司,内蒙古 呼伦贝尔 021000
西安石油大学,西安 710065
华能日照电厂,山东 日照 276800
收稿:2025-07-22,
修回:2025-08-11,
纸质出版:2026-04-25
移动端阅览
李志鹏, 姜楠, 孙国辉, 等. 基于多尺度熵-动态对齐融合法结合AFGRU的储能调频指令预测[J]. 电力电容器与无功补偿, 2026,47(2):37-44.
LI Zhipeng, JIANG Nan, SUN Guohui, et al. Prediction of Energy Storage Frequency Regulation Commands Based on Multi-scale Entropy Dynamic Alignment Fusion Method Combined with AFGRU[J]. Power Capacitor & Reactive Power Compensation, 2026, 47(2): 37-44.
李志鹏, 姜楠, 孙国辉, 等. 基于多尺度熵-动态对齐融合法结合AFGRU的储能调频指令预测[J]. 电力电容器与无功补偿, 2026,47(2):37-44. DOI: 10.14044/j.1674-1757.pcrpc.2026.02.004.
LI Zhipeng, JIANG Nan, SUN Guohui, et al. Prediction of Energy Storage Frequency Regulation Commands Based on Multi-scale Entropy Dynamic Alignment Fusion Method Combined with AFGRU[J]. Power Capacitor & Reactive Power Compensation, 2026, 47(2): 37-44. DOI: 10.14044/j.1674-1757.pcrpc.2026.02.004.
为提升电厂调频指令序列的预测精度,提出一种融合多尺度熵-动态对齐融合法(multi scale entropy dynamic alignment fusion,MSEDAF)与改进神经网络的组合预测模型。该方法首先将原始序列经VMD分解为6个子序列(intrinsic mode function,IMF),进而采用创新的多尺度熵-动态对齐融合法(multi scale entropy dynamic alignment fusion,MSEDAF)通过结合多尺度熵度量与动态时间规整的复合相似性计算,将规律性相似的子序列两两融合为3个综合分量(Fused);随后,设计自适应特征门控单元(adaptive featuregated recurrent unit,AFGRU)网络,其核心特点在于注意力机制和特征增强模块,对3个融合子序列并行预测并重构最终结果。基于内蒙古某电厂调频数据的实验表明:MSEDAF-VMD-AFGRU预测的MAE较标准GRU平均降低90.14%。该组合模型通过物理规律引导的特征融合与自适应门控机制,显著增强了复杂调频场景的预测精确度,为电网AGC控制系统提供了高精度支撑。
For improving the prediction accuracy of frequency regulation command sequences in power plants,a combined prediction model integrating multi cale entropy dynamic alignment fusion(MSEDAF) and improved neural networks is proposed. This method firstly decomposes the original sequence into 6 subsequences(Intrinsic Mode Function,IMF)through VMD,and then uses the innovative Multi scale entropy dynamic alignment fusion method(MSEDAF)to combine multi-scale entropy measurement with dynamic time regularization composite similarity calculation,and fuse the subsequences with similar regularity into 3 comprehensive components(Fused)pairwise.Then,an adaptive feature gated recurrent unit (AFGRU)network is designed,with its core features being attention mechanism and feature enhancement module,to predict and reconstruct the final results in parallel for the three fused subsequences. The experiments based on frequency regulation data from a power plant in Inner Mongolia show that the MAE predicted byMSEDAF-VMD-AFGRUis reduced by an average of 90.14% compared to the standard GRU. This combination model significantly enhances the prediction accuracy of complex frequency modulation scenarios through feature fusion guided by physical laws and adaptive gating mechanism,providing high precision support for the AGC control system of power grid.
黄策,燕云飞,沈迎,等.超容储能辅助火电机组调频的电气问题研究[J].电气技术,2022,23(8):103-108.
HUANG Ce,YAN Yunfei,SHEN Ying,et al. Research on electrical issues of frequency regulation of thermal power units assisted by super capacity energy storage [J]. Electrical Technology,2022,23(8):103-108.
殷悦,孙钢虎,兀鹏越,等.混合储能调频工程经济性提升策略研究及应用[J].热力发电,2024,53(10):81-89.
YIN Yue,SUN Ganghu,WUPengyue,etal. Research and application of economic enhancement strategies for hybrid energy storage frequency regulation engineering [J]. Thermal Power Generation,2024,53(10):81-89.
张江广,桂成中,段志,等.基于VMD-GRU-DTW组合模型的在役桥梁健康状态预测预警技术[J].防灾科技学院学报,2025,27(1):31-40.
ZHANG Jiangguang,GUI chengzhong,DUAN Zhi;et al. Prediction and early warning technology for the health status of in-service bridges based onVMD-GRU-DTWcombination model [J]. Journal of Disaster Prevention Science and Technology College,2025,27(1):31-40.
刘航源,陈伟涛,李远耀,等.基于EEMD-CNN-LSTM的新型综合模型在滑坡位移预测中的应用[J].地质力学学报,2024,30(4):633-646.
LIU Hangyuan,CHEN Weitao,LI Yuanyao,et al. Application of a new comprehensive model based onEEMD-CNNLSTM in landslide displacement prediction [J]. Journal of Geomechanics,2024,30(4):633-646.
徐冬梅,廖安栋,王文川.基于VMD-EEMD-CNN-LSTM混合模型的月径流预测[J].水利规划与设计,2023(7):57-63.
XU Dongmei,LIAO Andong,WANG Wenchuan,et al. Monthly runoff prediction based on VMD-EEMD-CNNLSTM hybrid model [J]. Water Resources Planning and Design,2023(7):57-63.
王凤,尚若冰.基于EEMD-CNN-LSTM的短期风电功率预测方法:CN115510739a [P]. 2022-12-23.
方明宽,宁静,陈春俊.高速列车横向蛇行失稳的EEMDCNN-LSTM预测方法[J].中国测试,2021,47(7):79-83.
FANG Mingkuan,NING Jing;CHEN Chunjun. EEMDCNN-LSTMprediction method for lateral instability of highspeed trains [J]. China Test,2021,47(7):79-83.
夏天,杨超,宋进良,等.基于动态事件触发复合输出一致性的直流微网电池储能系统分布式控制策略[J].电力电容器与无功补偿,2025,46(2):44-52.
XIA Tian,YANG Chao,SONG Jinliang,et al. Distributed control strategy for battery energy storage system in DC microgrid based on dynamic event-triggered composite output consistency[J]. Power Capacitor &Reactive Power Compensation,2025,46(2):44-52.
奚增辉,王卫斌,陆嘉铭.运用电力大数据构建企业污染指数及监测分析[J].西安工程大学学报2024,38(6):83-89.
XI Zenghui,WANG Weibin,LU Jiaming,Constructing enterprise pollution index and monitoring analysis using electricity big data[J]. Journal of Xi'an Polytechnic University,2024,38(6):83-89.
孟伟,俞斌,白隆,等.基于STGCN-Transformer的短期电力净负荷预测[J].中国测试,2025,51(6):160-169.
MENG Wei,YU Bin,BAI Long,et al. Wait short term electricity net load forecasting based on STGCN transform er [J]. China Test,2025,51(6):160-169
张启龙,陈湘萍.考虑风电消纳多向量流系统的遗传优化方法[J].电测与仪表,2023,60(3):115-121.
Zhang Qilong,Chen Xiangping. Genetic optimization method for multi-vector flow system considering wind power consumption[J]. Electrical Measurement & Instrumentation, 2023,60(3):115-121.
董明,李晓枫,杨章,等.基于数据驱动的分布式光伏发电功率预测方法研究进展[J].电网与清洁能源,2024, 40(1):8-17.
DONG Ming,LI Xiaofeng,YANG Zhang,et al. Research progress on data-driven prediction methods for distributed photovoltaic power generation [J]. Power System and Clean Energy,2024,40(1):8-17.
邓韦斯,卢斯煜,刘显茁,等.基于相空间重构和BiLSTM的风电功率短期预测[J].广东电力,2023,36(7):22-30.
DENG Weisi,LU Siyu,LIU Xianzhuo,et al. Short term wind power prediction based on phase space reconstruction and BiLSTM [J]. Guangdong Electric Power,2023,36(7):22-30
苏华英,林晨,张俨,等.基于改进AP聚类和双重注意力机制的区域级新能源超短期出力预测方法[J].广东电力,2025,38(3):8-17.
SU Huaying,LIN Chen,ZHANG Yan,et al. Regional level ultra short term output prediction method for new energy based on improved AP clustering and dual attention mechanism [J]. Guangdong Electric Power,2025,38(3):8-17
LIU B,ZHAO J B,HUANG Q,et al. Nonlinear virtual inertia control of WTGs for enhancing primary frequency response and suppressing drivetrain torsional oscillations[J]. IEEE Transactions on Power Systems,2021,36(5):4125-4136.
PU D G,YUE Z,TIM G,et al. Resilient secondary voltage control of islanded microgrids:an eskbf-based distributed fast terminal sliding mode control approach[J]. IEEE Transactions on Power Systems,2020,35(3):2319-2331.
DANNY O,SERGIO M. A simplified electro-mechanical model of a DFIG-based wind turbine for primary frequency control studies[J]. IEEE Latin America Transactions,2016, 14(8):3600-3608.
LEE J,JANG G,MULJADI E,et al. Stable short-term frequency support using adaptive gains for a DFIG-based wind power plant[J]. IEEE Transactions on Energy Conversion, 2016,31(3):1068-1080.
李志鹏,张宇翼,马益文,等.一种基于VMD分解与交换法的储能调频指令预测方法及系统:CN120222418a [P]. 2025-06-27.
李志鹏,王小辉,康祯,等.一种储能调频指令预测方法及系统:CN120165407a [P]. 2025-06-17.
朱海南,李丰硕,孙华忠,等.基于改进AlexNet-GRU深度学习网络的配电网短期负荷预测方法[J].电力电容器与无功补偿,2023,44(4):48-54.
ZHU Hainan,LI Fengshuo,SUN Huazhong,et al. Shortterm load forecasting method for distribution network based on improved AlexNet-GRU deep learning network [J]. Electric Power Capacitors and Reactive Power Compensation, 2023,44(4):48-54.
许道林,谢兵,伍毅,等.基于多层双向递归神经网络的短期电力负荷预测[J].电力电容器与无功补偿,2022,43 (2):96-104.
XU Daolin,XIE Bing,WU Yi,et al. Short-term power load forecasting based on multi-layer bidirectional recurrent neural network [J]. Electric Power Capacitors and Reactive Power Compensation,2022,43(2):96-104.
吴昌兵,李志鹏,兀鹏越,等.一种超级电容耦合锂电池的储能调频控制方法及系统:CN120200277a [P]. 2025-06-24.
杨沛豪,兀鹏越,王小辉,等.一种储能架构优化方法及系统:CN120105647a [P]. 2025-06-06.
李志鹏,张宇翼,马益文,等.基于运算改进型VMD的超级电容储能调频并网方法及系统:CN119834330b [P]. 2025-06-24.
0
浏览量
0
下载量
0
CSCD
0
CNKI被引量
关联资源
相关文章
相关作者
相关机构
陕公网安备 61010402000197