基于改进生成对抗网络的风光气象资源联合场景生成方法

doi:10.11930/j.issn.1004-9649.202410048

中国电力 ›› 2025, Vol. 58 ›› Issue (3): 183-192.DOI: 10.11930/j.issn.1004-9649.202410048

基于改进生成对抗网络的风光气象资源联合场景生成方法

姜通海¹(), 王峰¹(), 刘子琪¹(), 单帅杰²()

1. 中广核（山东）新能源投资有限公司，山东济南　250014
2. 山东大学电气工程学院，山东济南　250061

收稿日期:2024-10-15 出版日期:2025-03-28 发布日期:2025-03-26
作者简介:
姜通海（1970），男，高级工程师，从事多能互补源网荷储一体化集成与协同优化、基于人工智能技术的源网荷储分布式资源协同控制、高精度新能源功率预测驱动的源网荷储协调规划，E-mail：13940952916@163.com
王峰（1978），男，高级工程师，从事新能源场站开发建设及运营管理技术，E-mail：wangfengxuzhou@163.com
刘子琪（2000），女，工程师，从事新能源开发建设及运营管理，E-mail：1575903717@qq.com
单帅杰（1997），男，通信作者，博士研究生，从事新能源运行管理、电力经济、电力供需研究，E-mail：shanshuaijie@126.com
基金资助:
国家电网有限公司科技项目（4000-202355381A-2-3-XG）；国家自然科学基金资助项目（52177095）。

A Joint Scenario Generation Method for Wind-Solar Meteorological Resources Based on Improved Generative Adversarial Network

Tonghai JIANG¹(), Feng WANG¹(), Ziqi LIU¹(), Shuaijie SHAN²()

1. CGN (Shandong) New Energy Investment Co., Ltd., Jinan 250014, China
2. School of Electrical Engineering, Shandong University, Jinan 250061, China

Received:2024-10-15 Online:2025-03-28 Published:2025-03-26
Supported by:
This work is supported by Science and Technology Project of SGCC (No.4000-202355381A-2-3-XG) and National Natural Science Foundation of China (No.52177095).

摘要/Abstract

摘要：

随着可再生能源的大规模发展，风光气象资源的波动性和不确定性给电力系统安全稳定运行带来极大的影响。为支持电力系统规划与决策，克服新能源功率预测过程中气象资源的不确定性，提出了一种基于深度卷积生成对抗网络（deep convolutional generative adversarial networks，DCGAN）的风光气象资源联合场景生成方法。首先，根据风电、光伏场站的空间分布规律，将新能源场站空间分布与气象要素相关联；其次，利用DCGAN技术对风速-辐照度气象要素进行长时间序列场景模拟，实现区域范围的风光资源联合场景生成；然后，借助K-medoids方法消减冗余场景，保留聚类中心处的典型场景，进而直接评价场景生成效果，将气象要素转化为风电、光伏出力，对典型场景间接评价；最后，通过分析气象要素生成场景的直接评价与间接评价结果，检验场景生成算法的有效性。算例结果表明，基于DCGAN模型生成的典型场景能有效地反映区域空间风光分布规律，并在气象要素场景生成中具有较好的适用性。

关键词: 可再生能源, 联合场景生成, 场景消减, 深度学习

Abstract:

With the large-scale development of renewable energy, the volatility and uncertainty of wind and solar meteorological resources have a great impact on the safe and stable operation of the power system. To support the power system planning and decision-making, and overcome the uncertainty of meteorological resources in predicting new energy power, this paper proposes a wind-solar meteorological resource joint scenario generation method based on the deep convolutional generative adversarial networks (DCGAN). Firstly, based on the spatial distribution pattern of wind and photovoltaic power stations, the spatial distribution of new energy stations are associated with meteorological factors. Secondly, the DCGAN technology is used to simulate the long time series scenarios of wind speed-irradiance meteorological elements, thus achieving the joint scenario generation of wind- solar resources in a regional scope; Next, the K-medoids method is used to reduce the redundance scenarios and keep the typical scenarios at the cluster center, subsequently realizing the direct evaluation of the scenario generation effects. And then, the meteorological elements are converted into wind power and photovoltaic output, thus achieving the indirect evaluation of typical scenarios. Finally, the effectiveness of the scenario generation algorithm is tested through analyzing the direct and indirect evaluation results of the meteorological element generation scenarios. The case study results show that the typical scenarios generated based on the DCGAN model can effectively reflect the spatial distribution of wind-solar resources in the region, and the proposed method has good applicability in generating meteorological element scenarios.

Key words: renewable energy, joint scenario generation, scenario reduction, deep learning

姜通海, 王峰, 刘子琪, 单帅杰. 基于改进生成对抗网络的风光气象资源联合场景生成方法[J]. 中国电力, 2025, 58(3): 183-192.

Tonghai JIANG, Feng WANG, Ziqi LIU, Shuaijie SHAN. A Joint Scenario Generation Method for Wind-Solar Meteorological Resources Based on Improved Generative Adversarial Network[J]. Electric Power, 2025, 58(3): 183-192.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202410048

https://www.electricpower.com.cn/CN/Y2025/V58/I3/183

图/表 13

图 1 GAN结构

Fig.1 The basic structure of GAN

图 2 DCGAN结构

Fig.2 The basic structure of DCGAN

图 3 总体流程

Fig.3 The overall process

图 4 生成风速曲线

Fig.4 The wind speed curve

图 5 生成辐照度曲线

Fig.5 The solar irradiance curve

表 1 典型场景概率值

Table 1 Probability values for typical scenarios

场景序号	生成概率	场景序号	生成概率
1	0.1522	6	0.0836
2	0.0695	7	0.1677
3	0.1181	8	0.1742
4	0.0075	9	0.0409
5	0.0961	10	0.0902

图 6 风光联合场景功率曲线

Fig.6 The wind-solar joint scenario power curve

图 7 气象联合模型的典型场景出力情况

Fig.7 The typical scenario output of meteorological joint model

表 2 DCGAN方法场景生成离散性和极端性评价指标

Table 2 DCGAN method scenario generation discreteness and extreme evaluation indicators

场景生成方法	年分布偏差/%		年峰值偏差/%		资源变异系数/%
场景生成方法	风速	辐照度	风速	辐照度	风速均值	风速最大值	辐照度均值	辐照度最大值
传统	0.69	5.23	17.96	8.86	8.36	51.22	13.51	54.61
DCGAN	0.32	1.08	3.65	6.57	3.75	18.24	4.12	18.72

表 3 风光资源联合出力评价指标

Table 3 Joint output evaluation indicators of wind and solar resources

置信度/%	DCGAN		传统方法
置信度/%	覆盖率/%	功率区间/MW	覆盖率/%	功率区间/MW
100	97.42	390.32	98.04	562.48
95	92.31	292.78	97.38	434.25
90	86.75	222.71	94.29	315.08

表 4 模型计算效率评价指标

Table 4 Evaluation indicators of model computation efficiency

场景生成方法	计算时间/min	迭代次数
DCGAN	3.4	50
传统方法	7.3	50

图 8 概率密度函数曲线

Fig.8 Probability density function curve

图 9 经验累积分布函数曲线

Fig.9 Empirical cumulative distribution function curve

参考文献 25

1	LI M L, YANG M, YU Y X, et al. Short-term wind power forecast based on continuous conditional random field[J]. IEEE Transactions on Power Systems, 2024, 39 (1): 2185- 2197. DOI
2	胡源, 薛松, 杨素, 等. 综合能源背景下的配电网多场景规划[J]. 中国电力, 2021, 54 (4): 175- 184.
	HU Yuan, XUE Song, YANG Su, et al. Multi-scenario planning of distribution network in the context of integrated energy[J]. Electric Power, 2021, 54 (4): 175- 184.
3	薛禹胜, 雷兴, 薛峰, 等. 关于风电不确定性对电力系统影响的评述[J]. 中国电机工程学报, 2014, 34 (29): 5029- 5040.
	XUE Yusheng, LEI Xing, XUE Feng, et al. A review on impacts of wind power uncertainties on power systems[J]. Proceedings of the CSEE, 2014, 34 (29): 5029- 5040.
4	ZHANG Y M, ZHANG X, JI X Q, et al. Optimization of integrated energy system considering transmission and distribution network interconnection and energy transmission dynamic characteristics[J]. International Journal of Electrical Power & Energy Systems, 2023, 153, 109357.
5	任景, 高敏, 程松, 等. 面向新能源不确定性的西北电力电量平衡机制[J]. 中国电力, 2023, 56 (9): 66- 78.
	REN Jing, GAO Min, CHENG Song, et al. A balance method for power supply-demand adapting to high uncertainties of renewable energy in northwest power grid[J]. Electric Power, 2023, 56 (9): 66- 78.
6	李湃, 王伟胜, 刘纯, 等. 张北柔性直流电网工程新能源与抽蓄电站配置方案运行经济性评估[J]. 中国电机工程学报, 2018, 38 (24): 7206- 7214, 7447.
	LI Pai, WANG Weisheng, LIU Chun, et al. Economic assessment of Zhangbei VSC-based DC grid planning scheme with integration of renewable energy and pumped-hydro storage power station[J]. Proceedings of the CSEE, 2018, 38 (24): 7206- 7214, 7447.
7	曹慧秋, 徐箭, 洪敏, 等. 基于通用分布的风电功率出力动态场景生成方法[J]. 电测与仪表, 2018, 55 (24): 6- 10. DOI
	CAO Huiqiu, XU Jian, HONG Min, et al. A method of dynamic scenarios generation of wind power output based on versatile distribution[J]. Electrical Measurement & Instrumentation, 2018, 55 (24): 6- 10. DOI
8	MAKAROV Y V, ETINGOV P V, MA J, et al. Incorporating uncertainty of wind power generation forecast into power system operation, dispatch, and unit commitment procedures[J]. IEEE Transactions on Sustainable Energy, 2011, 2 (4): 433- 442. DOI
9	MA X Y, SUN Y Z, FANG H L. Scenario generation of wind power based on statistical uncertainty and variability[J]. IEEE Transactions on Sustainable Energy, 2013, 4 (4): 894- 904. DOI
10	李丹, 梁云嫣, 缪书唯, 等. 基于高斯混合聚类和改进条件变分自编码的多风电场功率日场景生成方法[J]. 中国电力, 2024, 57 (12): 17- 29.
	LI Dan, LIANG Yunyan, MIAO Shuwei, et al. Daily power scenario generation method for multiple wind farms based on Gaussian mixture clustering and improved conditional variational autoencoder[J]. Electric Power, 2024, 57 (12): 17- 29.
11	赵书强, 金天然, 李志伟, 等. 考虑时空相关性的多风电场出力场景生成方法[J]. 电网技术, 2019, 43 (11): 3997- 4004.
	ZHAO Shuqiang, JIN Tianran, LI Zhiwei, et al. Wind power scenario generation for multiple wind farms considering temporal and spatial correlations[J]. Power System Technology, 2019, 43 (11): 3997- 4004.
12	邹斌, 李冬. 基于有效容量分布的含风电场电力系统随机生产模拟[J]. 中国电机工程学报, 2012, 32 (7): 23- 31, 187.
	ZOU Bin, LI Dong. Power system probabilistic production simulation with wind generation based on available capacity distribution[J]. Proceedings of the CSEE, 2012, 32 (7): 23- 31, 187.
13	LEE D, BALDICK R. Future wind power scenario synthesis through power spectral density analysis[J]. IEEE Transactions on Smart Grid, 2014, 5 (1): 490- 500. DOI
14	PANDŽIĆ H, DVORKIN Y, QIU T, et al. Toward cost-efficient and reliable unit commitment under uncertainty[J]. IEEE Transactions on Power Systems, 2016, 31 (2): 970- 982. DOI
15	廖文龙, 杨德昌, 葛磊蛟, 等. 基于像素卷积生成网络的可再生能源场景生成[J]. 高电压技术, 2022, 48 (4): 1320- 1331.
	LIAO Wenlong, YANG Dechang, GE Leijiao, et al. Renewable energy scenario generation based on pixel convolutional generative networks[J]. High Voltage Engineering, 2022, 48 (4): 1320- 1331.
16	王守相, 陈海文, 李小平, 等. 风电和光伏随机场景生成的条件变分自动编码器方法[J]. 电网技术, 2018, 42 (6): 1860- 1869.
	WANG Shouxiang, CHEN Haiwen, LI Xiaoping, et al. Conditional variational automatic encoder method for stochastic scenario generation of wind power and photovoltaic system[J]. Power System Technology, 2018, 42 (6): 1860- 1869.
17	JIANG C M, MAO Y F, CHAI Y, et al. Scenario generation for wind power using improved generative adversarial networks[J]. IEEE Access, 2018, 6, 62193- 62203. DOI
18	李康平, 张展耀, 王飞, 等. 基于GAN场景模拟与条件风险价值的独立型微网容量随机优化配置模型[J]. 电网技术, 2019, 43 (5): 1717- 1725.
	LI Kangping, ZHANG Zhanyao, WANG Fei, et al. Stochastic optimization model of capacity configuration for stand-alone microgrid based on scenario simulation using GAN and conditional value at risk[J]. Power System Technology, 2019, 43 (5): 1717- 1725.
19	尹清青, 何涛, 吴馨, 等. 基于生成对抗网络的二次屏柜交直流侵入信号检测方法[J]. 东北电力大学学报, 2024, 44 (4): 21- 27.
	YIN Qingqing, HE Tao, WU Xin, et al. AC-DC intrusion signal detection method based on generative adversarial network for secondary panel cabinet[J]. Journal of Northeast Electric Power University, 2024, 44 (4): 21- 27.
20	李宝琴, 吴俊勇, 强子玥, 等. 基于改进DCGAN的电力系统暂态稳定增强型自适应评估[J]. 电力系统自动化, 2022, 46 (2): 73- 82.
	LI Baoqin, WU Junyong, QIANG Ziyue, et al. Enhanced adaptive assessment on transient stability of power system based on improved deep convolutional generative adversarial network[J]. Automation of Electric Power Systems, 2022, 46 (2): 73- 82.
21	宋宗耘, 张健, 牛东晓, 等. 基于模糊密度K-medoids聚类和NSGA-Ⅱ算法的绿色分时电价优化研究[J]. 智慧电力, 2019, 47 (3): 38- 45.
	SONG Zongyun, ZHANG Jian, NIU Dongxiao, et al. Green time of use electricity price optimization based on fuzzy density K-medoids algorithm and NSGA-Ⅱ[J]. Smart Power, 2019, 47 (3): 38- 45.
22	KONG X Y, XIAO J, LIU D H, et al. Robust stochastic optimal dispatching method of multi-energy virtual power plant considering multiple uncertainties[J]. Applied Energy, 2020, 279, 115707. DOI
23	ZHU J, ZHAO Z Y, ZHENG X R, et al. Time-series power forecasting for wind and solar energy based on the SL-transformer[J]. Energies, 2023, 16 (22): 7610. DOI
24	PURLU M, TURKAY B E. Optimal allocation of renewable distributed generations using heuristic methods to minimize annual energy losses and voltage deviation index[J]. IEEE Access, 2022, 10, 21455- 21474. DOI
25	杨皓然, 杨茂, 苏欣. 计及预测误差时空-条件相依特性的日内光伏出力区间预测方法[J]. 南方电网技术, 2023, 17 (2): 128- 136.
	YANG Haoran, YANG Mao, SU Xin. Intraday photovoltaic output interval prediction method considering the spatiotemporal-conditional dependence of prediction error[J]. Southern Power System Technology, 2023, 17 (2): 128- 136.

[1]	汤明润, 李若旸, 刘慕然, 程晓钰, 刘铫, 杨淑霞. 电力系统稳态下可再生能源大规模接入量预测[J]. 中国电力, 2025, 58(2): 126-132.
[2]	黄堃, 付明, 翟家祥, 华昊辰. 基于改进线性化ADMM的多微网经济运行分布式协调优化[J]. 中国电力, 2025, 58(2): 193-202.
[3]	鲁玲, 苑涛, 杨波, 李欣, 鲁洋, 蒲秋平, 张鑫. 计及㶲效率和多重不确定性的区域综合能源系统双层优化[J]. 中国电力, 2025, 58(1): 128-140.
[4]	邹小燕, 张瑞宏. 考虑政府干预的可再生能源与储能企业合作模式演化博弈研究[J]. 中国电力, 2025, 58(1): 153-163.
[5]	邵冲, 胡荣义, 余姣, 王明典. 考虑荷电与储氢状态的风光氢储系统动态控制仿真模型[J]. 中国电力, 2024, 57(7): 109-124.
[6]	胡旭, 安锐坚, 杜宇晨, 施啸寒, 王越. 欧洲—北非氢能协同发展研究[J]. 中国电力, 2024, 57(7): 151-162.
[7]	武群丽, 朱新宇. 可再生能源配额制下风光储联合参与现货市场的交易决策[J]. 中国电力, 2024, 57(4): 77-88.
[8]	王麒翔, 韩震焘, 梁毅. 基于博弈论均衡的风险防控经济调度分析[J]. 中国电力, 2024, 57(10): 69-77.
[9]	但扬清, 王蕾, 郑伟民, 武佳卉, 王晨轩, 余高旺. 高比例可再生能源接入背景下电网承载能力鲁棒提升策略[J]. 中国电力, 2023, 56(9): 104-111.
[10]	武震, 霍彦达, 张翼, 黄巨龙, 戴剑锋. 考虑新能源发电量季节性分布的省区级季节性储能需求测算方法[J]. 中国电力, 2023, 56(8): 40-47.
[11]	刘东, 张越, 皮俊波, 单连飞, 刘赫, 宋鹏程, 姜涛. 面向电网调控信息智能检索的知识图谱构建及应用[J]. 中国电力, 2023, 56(7): 78-84.
[12]	张鑫, 叶俊杰, 崔瑶, 黄鑫, 仲林林. 基于语义信息距离解耦的变电运维多类别缺陷图像检测[J]. 中国电力, 2023, 56(6): 209-218.
[13]	汤芳, 代红才, 张宁, 吴越, 薛美美, 陈睿. 能耗双控向碳排放双控转变影响分析及推进路径设计[J]. 中国电力, 2023, 56(12): 255-261.
[14]	周震震, 宋云海, 何宇浩, 王黎伟, 黄和燕, 何珏, 朱志航, 闫云凤. 基于分组查询注意力的可扩展电力人员行为分类方法[J]. 中国电力, 2023, 56(11): 77-85.
[15]	刘之琳, 许传龙, 郑海峰, 李江涛, 姚力, 刘青, 杜翼. 德国关停核电前后保供应促消纳经验分析[J]. 中国电力, 2023, 56(10): 145-152.

基于改进生成对抗网络的风光气象资源联合场景生成方法

A Joint Scenario Generation Method for Wind-Solar Meteorological Resources Based on Improved Generative Adversarial Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 25

相关文章 15

编辑推荐

Metrics

模态框（Modal）标题

基于改进生成对抗网络的风光气象资源联合场景生成方法

A Joint Scenario Generation Method for Wind-Solar Meteorological Resources Based on Improved Generative Adversarial Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献 25

相关文章 15

编辑推荐

Metrics