[1]刘志仁,杜云龙,颜全椿,等.基于多元周期性网格与云模型混合通道的功率预测研究[J].南京师大学报(自然科学版),2026,49(02):110-119.[doi:10.3969/j.issn.1001-4616.2026.02.011]
　Liu Zhiren,Du Yunlong,Yan Quanchun,et al.A Study of Power Prediction Based on a Hybrid Channel of Multivariate Periodic Grids and Cloud Models[J].Journal of Nanjing Normal University(Natural Science Edition),2026,49(02):110-119.[doi:10.3969/j.issn.1001-4616.2026.02.011]
点击复制

基于多元周期性网格与云模型混合通道的功率预测研究()

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
49
期数:
2026年02期
页码:
110-119
栏目:
计算机科学与技术
出版日期:
2026-04-10

文章信息/Info

Title:
A Study of Power Prediction Based on a Hybrid Channel of Multivariate Periodic Grids and Cloud Models
文章编号:
1001-4616(2026)02-0110-10
作者:
刘志仁1; 杜云龙2; 颜全椿3; 柴 赟2; 曹卫青3; 杨勤胜3
1.国网江苏省电力有限公司无锡供电分公司,江苏 无锡 214000
2.国网江苏省电力有限公司,江苏 南京 210098
3.江苏方天电力技术有限公司,江苏 南京 211000
Author(s):
Liu Zhiren2; Du Yunlong1; Yan Quanchun3; Chai Yun1; Cao Weiqing3; Yang Qinsheng3
1.State Grid Jiangsu Electric Power Co.,Ltd. Wuxi Power Supply Branch,Wuxi 214000,China
2.Jiangsu Electric Power Co.,Ltd.,Nanjing 210098,China
3.Jiangsu Frontier Power Technology Co.,Ltd.,Nanjing 211000,China
关键词:
新能源功率预测; 多元周期性网格; 卡尔曼滤波器; 云模型混合通道
Keywords:
new energy power prediction;  multivariate periodic grids;  Kalman filter;  cloud model hybrid channel
分类号:
O643/X703
DOI:
10.3969/j.issn.1001-4616.2026.02.011
文献标志码:
A
摘要:
随着新能源发电技术的广泛普及,功率预测已成为该领域不可或缺的研究方向. 然而,传统预测方法在面对不同时间尺度及多元因素影响时,其新能源功率预测模型的动态性能表现出明显局限性,致使预测精度难以达到理想水平. 针对该问题本研究提出一种结合多元周期性网格与云模型混合通道的功率预测技术研究,通过卡尔曼滤波器更新多元周期性网格,使其能实时适应复杂环境变化,精准捕捉时空动态特征,实现周期特征动态优化. 同时,利用云模型混合通道的隶属度函数与概率修正优化,融合多尺度信息并动态调整. 仿真结果表明,该方法有效提升了新能源功率预测在不同时空尺度下的准确性与稳定性,为新能源领域功率预测提供了更有效的解决方案. 在光伏功率预测中具有显著的推广意义,为电力系统调度优化、负荷管理和智能化控制提供有力支持,对促进光伏电站稳定运行和推动绿色能源发展具有重要的应用价值.
Abstract:
In the wake of the extensive dissemination and application of new energy generation technologies, power prediction has emerged as an essential research area within this domain. Nevertheless, conventional prediction methodologies exhibit conspicuous limitations in the dynamic capabilities of new energy power prediction models when confronted with diverse time scales and the impacts of multiple factors, thereby rendering the attainment of desired prediction accuracy a challenging feat. In response to this predicament, the present study proposes an innovative power prediction technique that amalgamates the multivariate periodic grid with the cloud model hybrid channel. Through the utilization of the Kalman filter, the multivariate periodic grid is continuously updated in real-time, enabling it to adeptly adapt to complex and fluctuating environmental conditions, precisely capture spatio-temporal dynamic characteristics, and effectuate the dynamic optimization of periodic features. Concurrently, the cloud model hybrid channel is enhanced by leveraging the membership function with probability correction to seamlessly integrate multi-scale information and effectuate dynamic adjustments. Simulation outcomes convincingly demonstrate that this method significantly augments the accuracy and stability of new energy power prediction across varying spatio-temporal scales, thereby proffering a more efficacious solution for power prediction within the new energy realm. This approach significantly enhances photovoltaic power forecasting, supporting power system dispatch optimization, load management, and intelligent control, while promoting the stable operation of photovoltaic plants and advancing green energy development.

参考文献/References:

[1]周孝信,赵强,张玉琼,等. “双碳”目标下我国能源电力系统发展趋势分析:绿电替代与绿氢替代[J]. 中国电机工程学报,2024,44(17):6707-6721.
[2]贠保记,张恩硕,张国,等. 考虑综合需求响应与“双碳”机制的综合能源系统优化运行[J]. 电力系统保护与控制,2022,50(22):11-19.
[3]陈运蓬,景超,白静波,等. 基于集成学习的新能源发电功率预测[J]. 太阳能学报,2024,45(6):412-421.
[4]周海锋,徐伟,沙立成,等. 考虑新能源不确定性的调度计划安全稳定校核方法[J]. 电力工程技术,2024,43(3):63-70.
[5]Wolff B,Kühnert J,Lorenz E,et al. Comparing support vector regression for PV power forecasting to a physical modeling approach using measurement,numerical weather prediction,and cloud motion data[J]. Solar Energy,2016,135:197-208.
[6]Dimd B D,Völler S,Midtgård O M,et al. Quantification of the impact of azimuth and tilt angle on the performance of a PV output power forecasting model for BIPVs[J]. IEEE Journal of Photovoltaics,2023,14(1):194-200.
[7]陈颖,孙荣富,吴志坚,等. 基于统计升尺度方法的区域风电场群功率预测[J]. 电力系统自动化,2013,37(7):1-5.
[8]Lydia M,Kumar S S,Selvakumar A I,et al. Linear and non-linear autoregressive models for short-term wind speed forecasting[J]. Energy Conversion and Management,2016,112:115-124.
[9]周文,孟良,杨正富,等. 基于时间序列的大型光伏电站发电短期预测[J]. 电源技术,2021,45(11):1490-1494.
[10]李逸航,肖辉,易纯,等. 基于STL分解和TPA机制的光伏功率区间预测[J]. 太阳能学报,2024,45(12):22-29.
[11]李楠,刘佳佳,赖心怡,等. 基于时间序列神经分层插值模型的光伏功率超短期多步预测[J]. 智慧电力,2024,52(4):69-77.
[12]叶林,裴铭,李卓,等. 风电和光伏发电功率联合预测与预调度框架[J]. 高电压技术,2024,50(9):3823-3836.
[13]陈殿昊,臧海祥,蒋雨楠,等. 基于多层次云图特征与宽度学习的超短期光伏功率预测[J]. 电力系统自动化,2024,48(22):131-139.
[14]贺亦琛,师长立,郭小强,等. 基于参数优化多核支持向量机的光伏功率预测算法[J]. 太阳能学报,2024,45(9):394-404.
[15]常青松,杨昭,杨熠辉,等. 基于相似日聚类的超短期光伏功率组合预测模型[J]. 热力发电,2023,52(11):123-131.
[16]张程珂,刘会灯,朱渝宁,等. 基于多特征分析提取的随机森林超短期光伏功率预测[J]. 电力需求侧管理,2023,25(6):50-56.
[17]刘志超,袁三男,唐万成. 基于BLSTM-随机森林的短期光伏发电输出功率预测[J]. 电源技术,2021,45(4):495-498.
[18]杨茂,王东,严干贵,等. 风电功率波动特性中的周期性研究[J]. 太阳能学报,2013,34(11):2020-2026.
[19]乔颖,孙荣富,丁然,等. 基于数据增强的分布式光伏电站群短期功率预测(二):网格化预测[J]. 电网技术,2021,45(6):2210-2218.
[20]战文华,车建峰,王勃,等. 基于网格化数值天气预报的区域光伏发电多输出功率预测方法[J]. 中国电力,2024,57(3):144-151.

备注/Memo

备注/Memo:
收稿日期:2025-08-14.
基金项目:国网江苏省电力有限公司科技项目资助项目(J2023116).
通讯作者:颜全椿,博士研究生,高级工程师,研究方向:新能源及储能数智应用. E-mail:yanquanchun@126.com
更新日期/Last Update: 2026-04-10