基于改进K-means聚类的风光发电场景划分

doi:10.12096/j.2096-4528.pgt.19090

发电技术 ›› 2020, Vol. 41 ›› Issue (6): 625-630.DOI: 10.12096/j.2096-4528.pgt.19090

基于改进K-means聚类的风光发电场景划分

宋学伟¹(), 刘玉瑶²

¹ 上海电机学院电气学院, 上海市浦东新区 201306
² 国网东营市垦利区供电公司, 山东省东营市 257000

收稿日期:2020-05-25 出版日期:2020-12-31 发布日期:2021-01-12
作者简介:宋学伟(1994), 男, 硕士, 从事负荷预测、主动配电网优化调度、孤岛运行研究, soongxuewei@163.com
刘玉瑶(1994), 女, 工程师, 从事可再生能源、电网计量工作
基金资助:
国家自然科学基金项目(51477099);上海市自然科学基金项目(15ZR1417300);上海市自然科学基金项目(14ZR1417200)

Wind and Photovoltaic Generation Scene Division Based on Improved K-means Clustering

Xuewei SONG¹(), Yuyao LIU²

¹ Department of Electrical Engineering, Shanghai Dian Ji University, Pudong New District, Shanghai 201306, China
² State Grid Dongying Kenli Power Supply Company, Dongying 257000, Shandong Province, China

Received:2020-05-25 Published:2020-12-31 Online:2021-01-12
Supported by:
National Natural Science Foundation of China(51477099);Shanghai Natural Science Foundation(15ZR1417300);Shanghai Natural Science Foundation(14ZR1417200)

摘要/Abstract

摘要：

针对可再生能源发电，尤其是风力、光伏发电的出力不确定性问题，结合改进后的K-means聚类方法对发电的状态进行场景划分。首先建立风力、光伏发电的不确定性模型，选用合适的概率密度函数进行拟合；之后结合密度聚类和提出的混合评价函数，对基本的K-means聚类算法进行改进，解决了算法的初始聚类中心和聚类个数难以选取的问题；然后运用改进后的K-means聚类对某地风力、光伏发电场景进行聚类划分，从而将不确定性问题转化成确定性问题。最后通过对场景划分的算例进行分析，验证了所提方法的工程实用性。

关键词: 风力发电, 光伏发电, 密度聚类, K-means聚类, 场景划分

Abstract:

In view of the uncertainty of power generation in renewable energy, especially wind power and photovoltaic power generation, the improved K-means clustering method was used to segment the state of power generation. Firstly, the uncertainty model of wind power and photovoltaic power generation was established, and the appropriate probability density function was used to fit. Then the basic K-means clustering algorithm was improved by combining density clustering and proposed hybrid evaluation function, to solve the problem that the initial clustering center and the number of clusters were difficult to select. The improved K-means clustering was used to cluster the wind and photovoltaic scenes in a certain place, thus transforming the uncertainty problem into a deterministic problem. Finally, the practicability of the proposed method was verified by analyzing an example of scenario division.

Key words: wind power generation, photovoltaic power generation, density clustering, K-means clustering, scenario division

中图分类号:

TK81
TM615

宋学伟, 刘玉瑶. 基于改进K-means聚类的风光发电场景划分[J]. 发电技术, 2020, 41(6): 625-630.

Xuewei SONG, Yuyao LIU. Wind and Photovoltaic Generation Scene Division Based on Improved K-means Clustering[J]. Power Generation Technology, 2020, 41(6): 625-630.

图/表 6

图1 基本K-means聚类流程

Fig.1 Basic K-means clustering process

图2 改进后的K-means聚类算法流程

Fig.2 Improved K-means clustering algorithm flow

表1 K-means聚类改进前后指标对比

Tab. 1 Comparison of indicators before and after K-means clustering improvement

指标	CH(+)	DB(-)	迭代次数	时间/s
K-means聚类	4 988.72	2.031 7	18	46
改进K-means聚类	5 366.87	1.726 3	6	22

表2 典型日场景概率

Tab. 2 Typical day scene probability

场景	概率
1	0.076 8
2	0.052 1
3	0.169 8
4	0.027 4
5	0.257 4
6	0.252 0
7	0.104 1
8	0.060 4

图3 典型日场景5

Fig.3 Typical day scene 5

图4 典型日场景6

Fig.4 Typical day scene 6

参考文献 15

1	白建华, 辛颂旭, 刘俊, 等. 中国实现高比例可再生能源发展路径研究[J]. 中国电机工程学报, 2015, 35 (14): 3699- 3705.
	BAI J H , XIN S X , LIU J , et al. Roadmap of realizing the high penetration renewable energy in China[J]. Proceedings of the CSEE, 2015, 35 (14): 3699- 3705.
2	张兴平, 刘文峰. 典型国家可再生能源政策演变研究[J]. 电网与清洁能源, 2018, 34 (10): 60- 68.
	ZHANG X P , LIU W F . The evolution of renewable energy policy in typical countries[J]. Power System and Clean Energy, 2018, 34 (10): 60- 68.
3	姜曼, 杨司玥, 刘定宜, 等. 中国各省可再生能源电力消纳量对碳排放的影响[J]. 电网与清洁能源, 2020, 36 (7): 87- 95.
	JIANG M , YANG S Y , LIU D Y , et al. Impacts of renewable electricity consumption on carbon dioxide emission in China's provinces[J]. Power System and Clean Energy, 2020, 36 (7): 87- 95.
4	马艺玮, 杨苹, 郭红霞, 等. 风-光-沼可再生能源分布式发电系统电源规划[J]. 电网技术, 2012, 36 (9): 9- 14.
	MA Y W , YANG P , GUO H X , et al. Power source planning of wind-PV-biogas renewable energy distributed generation system[J]. Power System Technology, 2012, 36 (9): 9- 14.
5	赵书强, 李志伟. 考虑可再生能源出力不确定性的多能源电力系统日前调度[J]. 华北电力大学学报(自然科学版), 2018, 45 (5): 1- 10.
	ZHAO S Q , LI Z W . Day-ahead scheduling of multi-energy power system considering renewable energy uncertain output[J]. Journal of North China Electric Power University (Nature Science), 2018, 45 (5): 1- 10.
6	于笑, 陈武晖. 风力发电并网系统次同步振荡研究[J]. 发电技术, 2018, 39 (4): 304- 312.
	YU X , CHEN W H . Review of subsynchronous oscillation induced by wind power generation integrated system[J]. Power Generation Technology, 2018, 39 (4): 304- 312.
7	庄雅妮, 杨秀媛, 金鑫城. 风光储联合发电运行技术研究[J]. 发电技术, 2018, 39 (4): 296- 303.
	ZHUANG Y N , YANG X Y , JIN X C . Study on operation technology of wind-PV-energy storage combined power generation[J]. Power Generation Technology, 2018, 39 (4): 296- 303.
8	张文秀, 韩肖清, 宋述勇, 等. 计及源-网-荷不确定性因素的马尔科夫链风电并网系统运行可靠性评估[J]. 电网技术, 2018, 42 (3): 762- 771.
	ZHANG W X , HAN X Q , SONG S Y , et al. Operational reliability evaluation of wind integrated power systems based on Markov chain considering uncertainty factors of source-grid-load[J]. Power System Technology, 2018, 42 (3): 762- 771.
9	赖昌伟, 黎静华, 陈博, 等. 光伏发电出力预测技术研究综述[J]. 电工技术学报, 2019, 34 (6): 1201- 1217.
	LAI C W , LI J H , CHEN B , et al. Review of photovoltaic power output prediction technology[J]. Transactions of China Electrotechnical Society, 2019, 34 (6): 1201- 1217.
10	谢富鹏, 姜文刚, 王茗倩. 光伏逐日发电装置变频率时控追踪控制策略研究[J]. 太阳能学报, 2019, 40 (4): 1011- 1020.
	XIE F P , JIANG W G , WANG M J . Study on control strategy for solar tracking system using variable tracking frequency[J]. Acta Energiae Solaris Sinica, 2019, 40 (4): 1011- 1020.
11	缪仁豪, 由世俊, 张欢, 等. 海岛生活舱风光柴互补发电系统优化设计[J]. 太阳能学报, 2018, 39 (8): 2147- 2154.
	MIU R H , YOU S J , ZHANG H , et al. Optimization design of island living quarters landscape and wood complementary power generation system[J]. Acta Energiae Solaris Sinica, 2018, 39 (8): 2147- 2154.
12	樊新东, 杨秀媛, 金鑫城. 风电场有功功率控制综述[J]. 发电技术, 2018, 39 (3): 268- 276.
	FAN X D , YANG X Y , JIN X C . An overview of active power control in wind farms[J]. Power Generation Technology, 2018, 39 (3): 268- 276.
13	韦仲康, 徐建飞, 于鹏. 考虑风力发电波动性的分布式发电可靠性规划[J]. 电力科学与工程, 2014, 30 (11): 31- 36.
	WEI Z K , XU J F , YU P . Reliability planning of distributed generation considering wind power fluctuations[J]. Electric Power Science and Engineering, 2014, 30 (11): 31- 36.
14	王群, 董文略, 杨莉. 基于Wasserstein距离和改进K-medoids聚类的风电/光伏经典场景集生成算法[J]. 中国电机工程学报, 2015, 35 (11): 2654- 2661.
	WANG Q , DONG W L , YANG L . A wind power/photovoltaic typical scenario set generation algorithm based on Wasserstein distance metric and revised K-medoids cluster[J]. Proceedings of the CSEE, 2015, 35 (11): 2654- 2661.
15	周开乐, 杨善林, 丁帅, 等. 聚类有效性研究综述[J]. 系统工程理论与实践, 2014, 34 (9): 2417- 2431.
	ZHOU L L , YANG S L , DING S , et al. On cluster validation[J]. Systems Engineering Theory & Practice, 2014, 34 (9): 2417- 2431.

[1]	肖瑶, 钮文泽, 魏高升, 崔柳, 杜小泽. 太阳能光伏/光热技术研究现状与发展趋势综述[J]. 发电技术, 2022, 43(3): 392-404.
[2]	吕清泉, 张珍珍, 马彦宏, 张健美, 高鹏飞, 蒋婷婷, 朱红路. 区域光伏发电出力特性分析研究[J]. 发电技术, 2022, 43(3): 413-420.
[3]	黄树帮, 陈耀, 金宇清. 碳中和背景下多通道特征组合超短期风电功率预测[J]. 发电技术, 2021, 42(1): 60-68.
[4]	朱月,沙俊民,赵静. 含不同机组风电场低电压穿越能力仿真研究[J]. 发电技术, 2020, 41(3): 328-333.
[5]	张哲旸,巨星,潘信宇,杨宇,徐超,杜小泽. 太阳能光伏-光热复合发电技术及其商业化应用[J]. 发电技术, 2020, 41(3): 220-230.
[6]	匡生,王蓓蓓. 考虑储能寿命和参与调频服务的风储联合运行优化策略[J]. 发电技术, 2020, 41(1): 73-78.
[7]	李相俊,许格健. 基于长短期记忆神经网络的风力发电功率预测方法[J]. 发电技术, 2019, 40(5): 426-433.
[8]	于航,刘阳,连魏魏,朱红路. 一种基于神经网络的硅基光伏组件运行温度在线软测量方法[J]. 发电技术, 2018, 39(6): 566-573.
[9]	仇梓峰,王爽心,李蒙. 基于无人机图像的风力发电机叶片缺陷识别[J]. 发电技术, 2018, 39(3): 277-285.
[10]	孔德同, 贾思远, 王天品, 刘庆超. 基于振动分析的风力发电机故障诊断方法[J]. 发电技术, 2017, 38(1): 54-58.

基于改进K-means聚类的风光发电场景划分

Wind and Photovoltaic Generation Scene Division Based on Improved K-means Clustering

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 15

相关文章 10

编辑推荐

Metrics

本文评价