基于堆叠式降噪自动编码器和深度神经网络的风电调频逐步惯性智能控制

doi:10.16183/j.cnki.jsjtu.2022.157

上海交通大学学报 ›› 2023, Vol. 57 ›› Issue (11): 1477-1491.doi: 10.16183/j.cnki.jsjtu.2022.157

所属专题：《上海交通大学学报》2023年“新型电力系统与综合能源”专题

• 新型电力系统与综合能源 • 上一篇下一篇

基于堆叠式降噪自动编码器和深度神经网络的风电调频逐步惯性智能控制

王亚伦¹, 周涛¹(), 陈中², 王毅³, 权浩¹

1.南京理工大学自动化学院, 南京 210094
2.东南大学电气工程学院,南京 210096
3.国网电力科学研究院有限公司智能电网保护和运行控制国家重点实验室,南京 211106

收稿日期:2022-05-13 修回日期:2022-09-08 接受日期:2022-11-04 出版日期:2023-11-28 发布日期:2023-12-01
通讯作者: 周涛,博士,讲师;E-mail:zhoutaonjust@njust.edu.cn.
作者简介:王亚伦(1997-),硕士生,从事电力系统频率稳定与控制研究.
基金资助:
江苏省自然科学基金青年项目(BK20220216);智能电网保护和运行控制国家重点实验室开放课题(SGNR0000KJJS2200305)

Stepwise Inertial Intelligent Control of Wind Power for Frequency Regulation Based on Stacked Denoising Autoencoder and Deep Neural Network

WANG Yalun¹, ZHOU Tao¹(), CHEN Zhong², WANG Yi³, QUAN Hao¹

1. School of Automation, Nanjing University of Science and Technology, Nanjing 210094, China
2. School of Electrical Engineering, Southeast University, Nanjing 210096, China
3. National Key Laboratory for Smart Grid Protection and Operation Control, State Grid Electric Power Research Institute Co., Ltd., Nanjing 211106, China

Received:2022-05-13 Revised:2022-09-08 Accepted:2022-11-04 Online:2023-11-28 Published:2023-12-01

摘要/Abstract

摘要：

风电调频的逐步惯性控制(SIC)策略在负荷波动后提供一个阶跃式功率增发,能够有效阻止系统频率下降,保障电网频率安全.但在其功率恢复阶段,容易出现二次频率跌落现象,需优化SIC以获得更好的调频效果.传统方法存在计算维度高和耗时较长的弊端,难以满足不同场景下快速提供最优控制效果的需求.为实现负荷扰动事件下风电调频的最优逐步惯性快速控制,引入深度学习算法,提出一种基于堆叠式降噪自动编码器(SDAE)和深度神经网络(DNN)的风电调频逐步惯性智能控制方法.首先,使用麻雀搜索算法获得最优参数,使用SDAE高效提取数据特征;随后,基于DNN对数据特征进行学习,并引入加速自适应矩估计优化网络参数,提升网络全局最优参数;最后,应用SDAE-DNN联合方法实现扰动事件后风电调频的逐步惯性在线控制.在IEEE 30节点测试系统中分别对单台风力机和风电场进行仿真分析,与传统方法、浅层反向传播神经网络及原始DNN所得结果对比发现,所提网络结构具有更优的预测精度和泛化能力,该方法能够实现良好的逐步惯性调频效果.

关键词: 逐步惯性控制, 二次频率跌落, 麻雀搜索算法, 堆叠式降噪自动编码器, 深度神经网络

Abstract:

Stepwise inertial control (SIC) provides a step-increase of power after load fluctuation, which can effectively prevent system frequency decline and ensure the safety of grid frequency. However, in the power recovery stage, secondary frequency drop (SFD) is easy to occur. Therefore, it is necessary to optimize SIC to obtain a better frequency regulation effect. The traditional method has the disadvantages of high calculation dimension and long consuming time, which is difficult to meet the requirements of providing the optimal control effect in different scenarios. In order to realize the optimal stepwise inertial fast control of wind power frequency regulation in load disturbance events, this paper introduces the deep learning algorithm and proposes a stepwise inertial intelligent control of wind power for frequency regulation based on stacked denoising autoencoder(SDAE) and deep neural network(DNN). First, sparrow search algorithm (SSA) is used to obtain the optimal parameters, and SDAE is used to extract the data features efficiently. Then, DNN is used to learn the data features, and the accelerated adaptive moment estimation is introduced to optimize the network parameters to improve the global optimal parameters of the network. Finally, the stepwise inertial online control of wind power frequency regulation after disturbance event is realized according to SDAE-DNN. The simulation analysis is conducted for a single wind turbine and a wind farm in the IEEE 30-bus test system. Compared with the results obtained by the traditional method, shallow BP neural network and original DNN network, it is found that the proposed network structure has a better prediction accuracy and generalization ability, and the proposed method can achieve a great effect of stepwise inertia frequency regulation.

Key words: stepwise inertial control (SIC), secondary frequency drop (SFD), sparrow search algorithm (SSA), stacked denoising autoencoder (SDAE), deep neural network (DNN)

中图分类号:

TM732

王亚伦, 周涛, 陈中, 王毅, 权浩. 基于堆叠式降噪自动编码器和深度神经网络的风电调频逐步惯性智能控制[J]. 上海交通大学学报, 2023, 57(11): 1477-1491.

WANG Yalun, ZHOU Tao, CHEN Zhong, WANG Yi, QUAN Hao. Stepwise Inertial Intelligent Control of Wind Power for Frequency Regulation Based on Stacked Denoising Autoencoder and Deep Neural Network[J]. Journal of Shanghai Jiao Tong University, 2023, 57(11): 1477-1491.

图/表 23

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

表1

图12

表2

图13

表3

表4

图14

表5

图15

表6

图16

表7

参考文献 30

[1]	卓振宇, 张宁, 谢小荣, 等. 高比例可再生能源电力系统关键技术及发展挑战[J]. 电力系统自动化, 2021, 45(9): 171-191.
	ZHUO Zhenyu, ZHANG Ning, XIE Xiaorong, et al. Key technologies and developing challenges of power system with high proportion of renewable energy[J]. Automation of Electric Power Systems, 2021, 45(9): 171-191.
[2]	国家能源局. 国家能源局: 我国可再生能源实现跨跃式发展—我国可再生能源发展有关情况介绍[J]. 中国电业, 2021(4): 6-9.
	National Energy Administration. National energy administration: China’s renewable energy realizes leap-forward development—Introduction of China’s renewable energy development[J]. China Electric Power, 2021(4): 6-9.
[3]	鲁宗相, 汤海雁, 乔颖, 等. 电力电子接口对电力系统频率控制的影响综述[J]. 中国电力, 2018, 51(1): 51-58.
	LU Zongxiang, TANG Haiyan, QIAO Ying, et al. The impact of power electronics interfaces on power system frequency control: A review[J]. Electric Power, 2018, 51(1): 51-58.
[4]	付媛, 王毅, 张祥宇, 等. 变速风电机组的惯性与一次调频特性分析及综合控制[J]. 中国电机工程学报, 2014, 34(27): 4706-4716.
	FU Yuan, WANG Yi, ZHANG Xiangyu, et al. Analysis and integrated control of inertia and primary frequency regulation for variable speed wind turbines[J]. Proceedings of the CSEE, 2014, 34(27): 4706-4716.
[5]	李军徽, 冯喜超, 严干贵, 等. 高风电渗透率下的电力系统调频研究综述[J]. 电力系统保护与控制, 2018, 46(2): 163-170.
	LI Junhui, FENG Xichao, YAN Gangui, et al. Survey on frequency regulation technology in high wind penetration power system[J]. Power System Protection & Control, 2018, 46(2): 163-170.
[6]	程志平, 张晗念, 徐亚利, 等. 风力发电调频策略研究现状分析[J]. 微电机, 2017, 50(10): 69-75.
	CHENG Zhiping, ZHANG Hannian, XU Yali, et al. Analysis on frequency control of wind turbines[J]. Micromotors, 2017, 50(10): 69-75.
[7]	YANG D J, LEE J, KANG Y C. Stepwise inertial control of a wind turbine generator to minimize a second frequency dip[J]. Journal of International Council on Electrical Engineering, 2016, 6(1): 153-159. doi: 10.1080/22348972.2016.1202396 URL
[8]	尹远, 卢继平, 刘钢, 等. 基于DFIG机组转子动能的风电场有功功率优化分配方法[J]. 电力系统保护与控制, 2012, 40(17): 127-132.
	YIN Yuan, LU Jiping, LIU Gang, et al. Active power distribution optimization of wind farm based on rotational kinetic energy of DFIG[J]. Power System Protection & Control, 2012, 40(17): 127-132.
[9]	ACKERMANN T, SÖDER L. Wind power in power systems: An introduction[M] //Wind power in power systems. Chichester, UK: John Wiley & Sons, Ltd., 2005: 25-51.
[10]	刘瑞. 双馈风机参与系统调频的二次跌落优化控制方法研究[D]. 北京: 华北电力大学, 2019.
	LIU Rui. Research on secondary drop optimized control method based on primary frequency control of doubly-fed wind turbines[D]. Beijing: North China Electric Power University, 2019.
[11]	WU Z P, GAO W Z, GAO T Q, et al. State-of-the-art review on frequency response of wind power plants in power systems[J]. Journal of Modern Power Systems & Clean Energy, 2018, 6(1): 1-16.
[12]	HAFIZ F, ABDENNOUR A. Optimal use of kinetic energy for the inertial support from variable speed wind turbines[J]. Renewable Energy, 2015, 80: 629-643. doi: 10.1016/j.renene.2015.02.051 URL
[13]	KANG M, KIM K, MULJADI E, et al. Frequency control support of a doubly-fed induction generator based on the torque limit[J]. IEEE Transactions on Power Systems, 2016, 31(6): 4575-4583. doi: 10.1109/TPWRS.2015.2514240 URL
[14]	BAO W Y, DING L, LIU Z F, et al. Analytically derived fixed termination time for stepwise inertial control of wind turbines—Part I: Analytical derivation[J]. International Journal of Electrical Power & Energy Systems, 2020, 121: 106120. doi: 10.1016/j.ijepes.2020.106120 URL
[15]	张旭, 陈云龙, 岳帅, 等. 风电参与电力系统调频技术研究的回顾与展望[J]. 电网技术, 2018, 42(6): 1793-1803.
	ZHANG Xu, CHEN Yunlong, YUE Shuai, et al. Retrospect and prospect of research on frequency regulation technology of power system by wind power[J]. Power System Technology, 2018, 42(6): 1793-1803.
[16]	刘洪波, 彭晓宇, 张崇, 等. 风电参与电力系统调频控制策略综述[J]. 电力自动化设备, 2021, 41(11): 81-92.
	LIU Hongbo, PENG Xiaoyu, ZHANG Chong, et al. Overview of wind power participating in frequency regulation control strategy for power system[J]. Electric Power Automation Equipment, 2021, 41(11): 81-92.
[17]	张怡, 张恒旭, 李常刚, 等. 深度学习在电力系统频率分析与控制中的应用综述[J]. 中国电机工程学报, 2021, 41(10): 3392-3406.
	ZHANG Yi, ZHANG Hengxu, LI Changgang, et al. Review on deep learning applications in power system frequency analysis and control[J]. Proceedings of the CSEE, 2021, 41(10): 3392-3406.
[18]	巩伟峥, 许凌, 姚寅. 计及风速分布与机组惯量转化不确定性的风电场可用惯量估计[J]. 上海交通大学学报, 2021, 55 (Sup.2): 51-59.
	GONG Weizheng, XU Ling, YAO Yin. Estimation of wind farm available inertia considering uncertainty of wind speed distribution and unit inertia transformation[J]. Journal of Shanghai Jiao Tong University, 2021, 55 (Sup.2): 51-59.
[19]	孙正龙, 李浩博, 刘铖, 等. 含虚拟惯量的双馈风电机组扭振阻尼特性分析与抑制方法研究[J]. 电网技术, 2021, 45(12): 4671-4683.
	SUN Zhenglong, LI Haobo, LIU Cheng, et al. Torsional oscillation damping characteristics and suppression methods of doubly-fed induction generator with virtual inertia[J]. Power System Technology, 2021, 45(12): 4671-4683.
[20]	乔颖, 郭晓茜, 鲁宗相, 等. 考虑系统频率二次跌落的风电机组辅助调频参数确定方法[J]. 电网技术, 2020, 44(3): 807-815.
	QIAO Ying, GUO Xiaoqian, LU Zongxiang, et al. Parameter setting of auxiliary frequency regulation of wind turbines considering secondary frequency drop[J]. Power System Technology, 2020, 44(3): 807-815.
[21]	ANDERSON P M, MIRHEYDAR M. A low-order system frequency response model[J]. IEEE Transactions on Power Systems, 1990, 5(3): 720-729. doi: 10.1109/59.65898 URL
[22]	ZHOU P, CHEN G, WANG M W, et al. Sediment classification of acoustic backscatter image based on stacked denoising autoencoder and modified extreme learning machine[J]. Remote Sensing, 2020, 12(22): 3762. doi: 10.3390/rs12223762 URL
[23]	XU K L, DARVE E. Solving inverse problems in stochastic models using deep neural networks and adversarial training[J]. Computer Methods in Applied Mechanics & Engineering, 2021, 384: 113976.
[24]	XUE J K, SHEN B. A novel swarm intelligence optimization approach: Sparrow search algorithm[J]. Systems Science & Control Engineering, 2020, 8(1): 22-34.
[25]	DOZAT T. Incorporating nesterov momentum into Adam[C]// International Conference on Learning Representations. San Juan, Puerto Rico: ICLR, 2016.
[26]	MA H, SHAHIDEHPOUR S M, MARWALI M K C. Transmission constrained unit commitment based on Benders decomposition[C]// Proceedings of the 1997 American Control Conference. Albuquerque, USA: IEEE, 1997: 2263-2267.
[27]	文云峰, 赵荣臻, 肖友强, 等. 基于多层极限学习机的电力系统频率安全评估方法[J]. 电力系统自动化, 2019, 43(1): 133-140.
	WEN Yunfeng, ZHAO Rongzhen, XIAO Youqiang, et al. Frequency safety assessment of power system based on multi-layer extreme learning machine[J]. Automation of Electric Power Systems, 2019, 43(1): 133-140.
[28]	刘陈续, 于桂兰. 基于神经网络的层状周期结构能量传输谱预测[J]. 上海交通大学学报, 2021, 55(1): 88-95.
	LIU Chenxu, YU Guilan. Prediction of energy transmission spectrum of layered periodic structures by neural networks[J]. Journal of Shanghai Jiao Tong University, 2021, 55(1): 88-95.
[29]	王同森, 程雪坤. 计及转速限值的双馈风机变下垂系数控制策略[J]. 电力系统保护与控制, 2021, 49(9): 29-36.
	WANG Tongsen, CHENG Xuekun. Variable droop coefficient control strategy of a DFIG considering rotor speed limit[J]. Power System Protection & Control, 2021, 49(9): 29-36.
[30]	王旭斌, 杜文娟, 王海风. 考虑锁相环动态的直驱风电机组虚拟惯性控制对电力系统小干扰稳定性影响[J]. 中国电机工程学报, 2018, 38(8): 2239-2252.
	WANG Xubin, DU Wenjuan, WANG Haifeng. Small-signal stability of power systems as affected by D-PMSG virtual inertia control considering PLL dynamics[J]. Proceedings of the CSEE, 2018, 38(8): 2239-2252.

预测网络	层数	算法	输入/输出层神经元个数	隐藏层神经元个数
SDAE-DNN	3+4	ADAM/ NADAM	3, 3/3, 2	9, 27, 18/ 6, 12, 6, 4
DNN	4	ADAM	3, 2	6, 12, 6, 4
BP	1	SGD	3, 2	12

预测网络	t/s	e_MSE
预测网络	t/s	ΔP_up	ΔT_up
SDAE-DNN	0.238	0.0025	0.0019
DNN	0.187	0.0081	0.0063
BP	0.153	0.0415	0.0347

预测网络	t/s	e_MSE
预测网络	t/s	ΔP_up	ΔT_up
SDAE-DNN	0.179	0.0093	0.0079
DNN	0.143	0.0171	0.0134

v_w/s	风电占比/%	负荷扰动量(p.u.)	ΔP_up (p.u.)	ΔT_up/s	f_nadir/Hz	最大RoCoF/(Hz·s^-1)
4	5	0.01	0.0026	8.32	49.9676	0.0258
4	5	0.03	0.0076	8.40	49.9027	0.0707
4	5	0.05	0.0127	8.41	49.8382	0.1079
5	10	0.05	0.0124	8.09	49.8346	0.1143
5	30	0.05	0.0130	7.29	49.8285	0.1449
5	50	0.05	0.0137	6.41	49.8208	0.1982
6	30	0.05	0.0126	7.22	49.8267	0.1458
8	30	0.05	0.0119	7.09	49.8234	0.1473
10	30	0.05	0.0112	7.02	49.8201	0.1489

参数	基于SDAE和DNN的SIIC	基于FTT的SIC	虚拟惯性控制	下垂控制	风电不参与调频
f_nadir/Hz	49.8989	49.8905	49.8768	49.8739	49.8611
最大RoCoF/(Hz·s^-1)	0.0860	0.0885	0.1037	0.1036	0.1041

基于堆叠式降噪自动编码器和深度神经网络的风电调频逐步惯性智能控制

Stepwise Inertial Intelligent Control of Wind Power for Frequency Regulation Based on Stacked Denoising Autoencoder and Deep Neural Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 23

参考文献 30

相关文章 3

编辑推荐

Metrics

本文评价

参数	SDAE-DNN	DNN	BP
f_nadir/Hz	49.8989	49.8983	49.8971
最大RoCoF/(Hz·s^-1)	0.0860	0.0867	0.0877

参数	基于SDAE和 DNN的SIIC	基于DNN的 SIIC	基于BP的 SIIC	基于FTT的 SIC	虚拟惯性控制	下垂控制	风电不参与调频
f_nadir/Hz	49.8607	49.8589	49.8563	49.8520	49.8351	49.8317	49.8174
最大RoCoF/(Hz·s^-1)	0.1100	0.1108	0.1120	0.1140	0.1294	0.1293	0.1298

[1]	刘宇, 雷雪梅. 融合MobileNetV3特征的结构化剪枝方法[J]. 上海交通大学学报, 2023, 57(9): 1203-1213.
[2]	熊磊, 苗雨润, 范新舟, 姚晔. 一种利用改进麻雀搜索算法的中央空调系统节能控制方法[J]. 上海交通大学学报, 2023, 57(4): 495-504.
[3]	郑德重, 杨媛媛, 谢哲, 倪扬帆, 李文涛. 基于Gaussian混合的距离度量学习数据划分方法[J]. 上海交通大学学报, 2021, 55(2): 131-140.