基于数据学习的新能源高渗透电网频率风险评估

doi:10.12096/j.2096-4528.pgt.20105

发电技术 ›› 2021, Vol. 42 ›› Issue (1): 40-47.DOI: 10.12096/j.2096-4528.pgt.20105

基于数据学习的新能源高渗透电网频率风险评估

温佳鑫¹(), 卜思齐¹^,*(), 陈麒宇²(), 周博文³()

¹ 香港理工大学电机工程学系, 香港特别行政区九龙区 999077
² 中国电力科学研究院有限公司, 北京市海淀区 100192
³ 东北大学信息科学与工程学院, 辽宁省沈阳市 110819

收稿日期:2020-09-30 出版日期:2021-02-28 发布日期:2021-03-12
通讯作者: 卜思齐
作者简介:卜思齐(1984)，男，博士，副教授，博士生导师，主要研究方向为电力电子化和智能信息化电力系统稳定控制分析与运行规划，本文通信作者，siqi.bu@polyu.edu.hk
温佳鑫(1989)，男，博士研究生，主要研究方向为新能源扰动下的大电网频率风险评估，Jiaxin.wen@connect.polyu.hk
陈麒宇(1986)，男，博士，高级工程师，主要研究方向为新能源及电力系统分析与规划，chenqiyu@epri.sgcc.com.cn
周博文(1987)，男，博士，讲师，主要研究方向为电力系统稳定与控制、电动汽车与电网互动、储能、需求响应、新能源、能源互联网等，zhoubowen@ise.neu.edu.cn
基金资助:
国家自然科学基金(51807171);香港研究资助局基金(15200418);香港研究资助局基金(15219619);广东省自然科学基金(2019A1515011226)

Data Learning-based Frequency Risk Assessment in a High-penetrated Renewable Power System

Jiaxin WEN¹(), Siqi BU¹^,*(), Qiyu CHEN²(), Bowen ZHOU³()

¹ Department of Electrical Engineering, The Hong Kong Polytechnic University, Kowloon District, Hong Kong S. A. R. 999077, China
² China Electric Power Research Institute, Haidian District, Beijing 100192, China
³ College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning Province, China

Received:2020-09-30 Published:2021-02-28 Online:2021-03-12
Contact: Siqi BU
Supported by:
National Natural Science Foundation of China(51807171);Hong Kong Research Grants Council(15200418);Hong Kong Research Grants Council(15219619);Natural Science Foundation of Guangdong Province(2019A1515011226)

摘要/Abstract

摘要：

新能源正在逐步代替传统发电厂为用户提供电能，但同时也为电网的安全运行带来了潜在的风险。因此，在规划阶段需要全面地对最大频率偏差越线风险进行概率评估。基于蒙特卡罗仿真（Monte Carlo simulation，MCS）的规划方法效率很低，而人工神经网络（artificial neural network，ANN）可以通过对数据的学习做出快速有效的预测。为此，提出一种基于MCS-ANN的区域频率概率评估方法，以实现对区域最大频率偏差越线风险的快速评估。首先，产生大量的随机扰动，仅对小部分扰动进行仿真；然后将这部分数据送入ANN进行训练，并将剩余的大部分扰动送入训练好的ANN进行输出预测；重复以上训练和预测的过程，将多次预测结果的平均值作为最终的预测输出，得到各个风险区间的概率分布情况。最后，在IEEE 10机39节点的系统上验证了所提方法的有效性。

关键词: 新能源, 电网安全, 频率风险评估, 蒙特卡罗仿真(MCS), 神经网络(ANN)

Abstract:

Renewable energy is gradually replacing traditional power plants to provide electricity for users, but it also brings potential risks to the safe operation of the power grid. Thus, in the planning stage, it is necessary to comprehensively evaluate the probability of violation risk of maximum frequency deviation. Planning method based on Monte Carlo simulation (MCS) is inefficient, while artificial neural network (ANN) can make fast and effective prediction by learning data. Therefore, this paper proposed an MCS-ANN algorithm to realize the rapid assessment of violation risk of regional maximum frequency deviation. Firstly, a large number of stochastic disturbances were generated, and only a small part of disturbances were used for MCS. Then, these data were sent to the neural network for training, and most of the remaining disturbances were sent to the trained neural network for output prediction. The above training and prediction processes were repeated. The average of multiple prediction results was used as the final prediction output, and the probability distribution of each risk interval was obtained. Finally, the effectiveness of the proposed MCS-ANN algorithm was verified on IEEE 10-machine 39-node system.

Key words: renewable energy, power grid security, frequency risk assessment, Monte Carlo simulation (MCS), artificial neural network (ANN)

中图分类号:

TK81

温佳鑫, 卜思齐, 陈麒宇, 周博文. 基于数据学习的新能源高渗透电网频率风险评估[J]. 发电技术, 2021, 42(1): 40-47.

Jiaxin WEN, Siqi BU, Qiyu CHEN, Bowen ZHOU. Data Learning-based Frequency Risk Assessment in a High-penetrated Renewable Power System[J]. Power Generation Technology, 2021, 42(1): 40-47.

图/表 16

图1 用于不同规划目标下的风能光能随机模型

Fig.1 Stochastic models of wind energy and solar energy for different planning objectives

图2 风速与风功率的转换关系

Fig.2 The relationship between wind speed and wind power

图3 扰动下的多机频率响应

Fig.3 Frequency responses of multi-machines after disturbances

图4 神经元结构

Fig.4 Structure of a neural network

图5 神经网络结构图

Fig.5 Structure of ANN

图6 基于MCS-ANN的区域频率概率评估流程

Fig.6 Regional frequency probability assessment process based on MCS-ANN

图7 含有3台风机的IEEE 10机39节点测试系统

Fig.7 IEEE 10-machine 39-node test system with 3 wind farms

图8 系统最大频率偏差的概率分布

Fig.8 Histogram of system maximum frequency deviation error distribution

图9 Area 1最大频率偏差误差分布直方图

Fig.9 Histogram of Area 1 maximum frequency deviation error distribution

图10 Area 2最大频率偏差误差分布直方图

Fig.10 Histogram of Area 2 maximum frequency deviation error distribution

图11 Area 3最大频率偏差误差分布直方图

Fig.11 Histogram of Area 3 maximum frequency deviation error distribution

表1 系统最大频率偏差的概率分布

Tab. 1 Probabilistic distribution of system maximum frequency deviation %

最大频率偏差/Hz	MCS-ANN	MCS	绝对误差
< 49.5	1.74	1.80	0.06
[49.5, 49.8)	21.74	21.04	0.70
[49.8, 50.2]	63.64	64.18	0.54
(50.2, 50.5]	12.88	12.98	0.10
> 50.5	0.00	0.00	0.00

表2 Area 1最大频率偏差的概率分布

Tab. 2 Probabilistic distribution of Area 1 maximum frequency deviation %

最大频率偏差/Hz	MCS-ANN	MCS	绝对误差
< 49.5	1.84	1.88	0.04
[49.5, 49.8)	22.37	22.04	0.33
[49.8, 50.2]	62.57	62.78	0.21
(50.2, 50.5]	13.22	13.30	0.08
> 50.5	0.00	0.00	0.00

表3 Area 2最大频率偏差的概率分布

Tab. 3 Probabilistic distribution of Area 2 maximum frequency deviation %

最大频率偏差/Hz	MCS-ANN	MCS	绝对误差
< 49.5	1.75	1.80	0.05
[49.5, 49.8)	21.85	21.22	0.63
[49.8, 50.2]	63.57	63.96	0.39
(50.2, 50.5]	12.83	13.02	0.19
> 50.5	0.00	0.00	0.00

表4 Area 3最大频率偏差的概率分布

Tab. 4 Probabilistic distribution of Area 3 maximum frequency deviation %

最大频率偏差/Hz	MCS-ANN	MCS	绝对误差
< 49.5	1.78	1.84	0.06
[49.5, 49.8)	21.87	21.36	0.51
[49.8, 50.2]	63.27	63.64	0.37
(50.2, 50.5]	13.07	13.16	0.09
> 50.5	0.00	0.00	0.00

表5 MCS-ANN与MCS方法的计算时间对比

Tab. 5 Comparison of computational time between MCS-ANN and MCS

方法	计算时间/s
MCS-ANN	714.4=700(100次MCS)+14.4(10次ANN训练与预测)
MCS	34 998.7(5 000次MCS)

参考文献 32

1	LI T , LI A , GUO X P . The sustainable development-oriented development and utilization of renewable energy industry: a comprehensive analysis of MCDM methods[J]. Energy, 2020, 212, 118694. DOI
2	潘旭东, 黄豫, 唐金锐, 等. 新能源发电发展的影响因素分析及前景展望[J]. 智慧电力, 2019, 47 (11): 41- 47.
	PAN X D , HUANG Y , TANG J R , et al. Influencing factors and prospects for development of renewable energy power generation[J]. Smart Power, 2019, 47 (11): 41- 47.
3	LU C , FENG Y W , FEI C W , et al. Improved decomposed-coordinated kriging modeling strategy for dynamic probabilistic analysis of multicomponent structures[J]. IEEE Transactions on Reliability, 2020, 69 (2): 440- 457. DOI
4	武志军. 综合能源服务下新能源集约化管控评价方法[J]. 华电技术, 2019, 41 (12): 68- 71.
	WU Z J . Evaluation method on new energy intensive management for integrated energy services[J]. Huadian Technology, 2019, 41 (12): 68- 71.
5	杜文娟, 卜思齐, 王海风. 考虑并网风电随机波动的电力系统小干扰概率稳定性分析[J]. 中国电机工程学报, 2011, 31 (S1): 7- 11.
	DU W J , BU S Q , WANG H F . Effect of stochastic variation of grid-connected wind generation on power system small-signal probabilistic stability[J]. The Proceedings of the CSEE, 2011, 31 (S1): 7- 11.
6	KROPOSKI B . Integrating high levels of variable renewable energy into electric power systems[J]. Journal of Modern Power Systems and Clean Energy, 2017, (5): 831- 837. DOI
7	李克勋, 宗明珠, 魏高升. 地热能及与其他新能源联合发电综述[J]. 发电技术, 2020, 41 (1): 79- 87.
	LI K X , ZONG M Z , WEI G S . Overview of geothermal power generation and joint power generation with other new energy sources[J]. Power Generation Technology, 2020, 41 (1): 79- 87.
8	朱罡, 王茂春. 含新能源发电的电力系统状态估计研究[J]. 电网与清洁能源, 2020, 36 (4): 32- 40.
	ZHU G , WANG M C . Review on state estimation for power system containing renewable energy generation[J]. Power System and Clean Energy, 2020, 36 (4): 32- 40.
9	LI R , GENG H , YANG G . Fault ride-through of renewable energy conversion systems during voltage recovery[J]. Journal of Modern Power Systems and Clean Energy, 2016, (4): 28- 39. DOI
10	段建民, 王志新, 王承民. 基于电网聚合模型的新能源接纳能力评价与分析[J]. 分布式能源, 2019, 4 (5): 42- 49.
	DUAN J M , WANG Z X , WANG C M . Evaluation and analysis of the capability of new energy integration based on grid aggregation mode[J]. Distributed Energy, 2019, 4 (5): 42- 49.
11	李更丰, 邱爱慈, 黄格超, 等. 电力系统应对极端事件的新挑战与未来研究展望[J]. 智慧电力, 2019, 47 (8): 1- 11.
	LI G F , QIU A C , HUANG G C , et al. New challenges and future research prospects in power system against to extreme events[J]. Smart Power, 2019, 47 (8): 1- 11.
12	KUNDUR P . Power system stability and control[M]. McGraw-Hill New York: Spring, 2013.
13	马世英, 王青. 大规模新能源集中外送系统源网协调风险及仿真评估[J]. 发电技术, 2018, 39 (2): 112- 117.
	MA S Y , WANG Q . Risk and simulation evaluation of large-scale centralized new energy sending system[J]. Power Generation Technology, 2018, 39 (2): 112- 117.
14	SHI Q , LI F , CUI H . Analytical method to aggregate multi-machine SFR model with applications in power system dynamic studies[J]. IEEE Transactions on Power Systems, 2018, 33 (6): 6355- 6367. DOI
15	BU S , WEN J , LI F . A generic framework for analytical probabilistic assessment of frequency stability in modern power system operational planning[J]. IEEE Transactions on Power Systems, 2019, 34 (5): 3973- 3976. DOI
16	WEN J , BU S , ZHOU B , et al. A fast-algorithmic probabilistic evaluation on regional rate of change of frequency (RoCoF) for operational planning of high renewable penetrated power systems[J]. Energies, 2020, 13, 2780. DOI
17	FRERIS L , LNFIELD D . Renewable energy in power system[M]. New York: Wiley, 2008.
18	BERIZZI A , BOVO C , DELFANTI M , et al. A Monte Carlo approach for TTC evaluation[J]. IEEE Transactions on Power Systems, 2007, 22 (2): 735- 743. DOI
19	HEYDT G T , GRAF T J . Distribution system reliability evaluation using enhanced samples in a Monte Carlo approach[J]. IEEE Transactions on Power Systems, 2010, 25 (4): 2006- 2008. DOI
20	AHMADYAR A , RIAZ A , VERBIC G , et al. A framework for assessing renewable integration limits with respect to frequency performance[J]. IEEE Transactions on Power Systems, 2018, 33 (4): 4444- 4453. DOI
21	LOPEZ-GARCIA T B , CORONADO-MENDOZA A , DOMÍNGUEZ-NAVARRO J A . Artificial neural networks in microgrids: a review[J]. Engineering Applications of Artificial Intelligence, 2020, 95, 103894. DOI
22	MUYEEN S M , HASANIEN H M , AL-DURRA A . Transient stability enhancement of wind farms connected to a multi-machine power system by using an adaptive ANN-controlled SMES[J]. Energy Conversion and Management, 2014, 78, 412- 420. DOI
23	GUIJO-RUBIO D , DURÁN-ROSAL A M , GUTIÉRREZ P A , et al. Evolutionary artificial neural networks for accurate solar radiation prediction[J]. Energy, 2020, 210, 118374. DOI
24	ZHANG X S , YANG B , YU T , et al. Dynamic surrogate model based optimization for MPPT of centralized thermoelectric generation system under heterogeneous temperature differences[J]. IEEE Transactions on Energy Conversion, 2020, 35 (2): 966- 976. DOI
25	BU S Q , DU W , WANG H F , et al. Probabilistic analysis of small-signal stability of large-scale power systems as affected by penetration of wind generation[J]. IEEE Transactions on Power Systems, 2012, 27 (2): 762- 770. DOI
26	王成山, 郑海峰, 谢莹华, 等. 计及分布式发电的配电系统随机潮流计算[J]. 电力系统自动化, 2005, 24, 39- 44.
	WANG C S , ZHENG H F , XIE Y H , et al. Probabilistic power flow containing distributed generation in distribution system[J]. Automation of Electric Power Systems, 2005, 24, 39- 44.
27	KENDALL M . Kendall's advanced theory statistics[M]. New York: Oxford University Press, 1987.
28	ANDERSON P M , FOUAD A A . Power system control and stability[M]. Ames, IA: Iowa State University Press, 1977.
29	National Electricity Transmission System Security and Quality of Supply Standards[S].UK: National Grid ESO, 2014.
30	AGATONOVIC-KUSTRIN S , BERESFORD R . Basic concepts of artificial neural network (ANN) modeling and its application in pharmaceutical research[J]. Journal of Pharmaceutical and Biomedical Analysis, 2000, 22 (5): 717- 727. DOI
31	ROGERS G . Power system oscillations[M]. Norwell, MA: Kluwer, 2000.
32	CHOW J H , CULLUM J , WILLOUGHBY R A . A sparsity-based technique for identifying slow-coherent areas in large power systems[J]. IEEE Transactions on Power Apparatus and Systems, 1984, 103 (3): 463- 473.

[1]	刘洪波, 刘珅诚, 盖雪扬, 刘永发, 阎禹同. 高比例新能源接入的主动配电网规划综述[J]. 发电技术, 2024, 45(1): 151-161.
[2]	陈皓勇, 黄宇翔, 张扬, 王斐, 周亮, 汤君博, 吴晓彬. 基于“三流分离-汇聚”的虚拟电厂架构设计[J]. 发电技术, 2023, 44(5): 616-624.
[3]	郁海彬, 张煜晨, 刘扬洋, 陆增洁, 翁锦德. 碳交易机制下多主体虚拟电厂参与电力市场的优化调度竞标策略[J]. 发电技术, 2023, 44(5): 634-644.
[4]	李建林, 邵晨曦, 张则栋, 梁忠豪, 曾飞. 氢能产业政策及商业化模式分析[J]. 发电技术, 2023, 44(3): 287-295.
[5]	杨锡勇, 张仰飞, 林纲, 张玉卓, 安允展, 杨昊天. 考虑需求响应的源-荷-储多时间尺度协同优化调度策略[J]. 发电技术, 2023, 44(2): 253-260.
[6]	肖洋, 李志强, 程林, 汤磊, 夏潮, 梁英, 宋锐, 王东阳, 李贺文. 西北电网集中式调相机AVC综合协调控制策略[J]. 发电技术, 2023, 44(2): 270-279.
[7]	印欣, 张锋, 阿地利·巴拉提, 常喜强, 陈武晖, 李长军, 李雪明, 袁少伟. 新型电力系统背景下电热负荷参与实时调度研究[J]. 发电技术, 2023, 44(1): 115-124.
[8]	董光德, 李道明, 陈咏涛, 马兴, 付昂, 穆钢, 肖白. 基于粒子群优化与卷积神经网络的电能质量扰动分类方法[J]. 发电技术, 2023, 44(1): 136-142.
[9]	郝艺博, 杜锡力, 李笑竹, 陈来军. 考虑储能性能差异的新能源场站群共享储能交易模式[J]. 发电技术, 2022, 43(5): 687-697.
[10]	郑俊生, 吕心荣, 郑剑平. 锂离子电容器性能分析及其应用[J]. 发电技术, 2022, 43(5): 775-783.
[11]	余潇, 卜广全, 王姗姗. 风电经柔直孤岛送出交流暂态过电压抑制策略研究[J]. 发电技术, 2022, 43(4): 618-625.
[12]	刘建伟, 李学斌, 刘晓鸥. 有源配电网中分布式电源接入与储能配置[J]. 发电技术, 2022, 43(3): 476-484.
[13]	房方, 梁栋炀, 刘亚娟, 胡阳, 刘吉臻. 海上风电智能控制与运维关键技术[J]. 发电技术, 2022, 43(2): 175-185.
[14]	朱凯, 张艳红. “双碳”形势下电力行业氢能应用研究[J]. 发电技术, 2022, 43(1): 65-72.
[15]	周保中, 刘敦楠, 张继广, 李忆, 徐尔丰, 毕圣. “风光火一体化”多能互补项目优化配置研究[J]. 发电技术, 2022, 43(1): 10-18.

基于数据学习的新能源高渗透电网频率风险评估

Data Learning-based Frequency Risk Assessment in a High-penetrated Renewable Power System

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

图/表 16

参考文献 32

相关文章 15

编辑推荐

Metrics