基于卷积神经网络的新型电力系统频率特性预测方法

doi:10.16183/j.cnki.jsjtu.2023.071

上海交通大学学报 ›› 2024, Vol. 58 ›› Issue (10): 1500-1512.doi: 10.16183/j.cnki.jsjtu.2023.071

• 新型电力系统与综合能源 • 上一篇下一篇

基于卷积神经网络的新型电力系统频率特性预测方法

陆文安¹, 朱清晓¹, 李兆伟², 刘辉³, 余一平¹()

1.河海大学能源与电气学院,南京 211100
2.南瑞集团(国网电力科学研究院)有限公司,南京 211106
3.国网安徽省电力有限公司,合肥 230022

收稿日期:2023-03-03 修回日期:2023-05-08 接受日期:2023-05-26 出版日期:2024-10-28 发布日期:2024-11-01
通讯作者: 余一平,教授;E-mail: yyiping@hhu.edu.cn.
作者简介:陆文安(1998—),硕士生,研究方向为新型电力系统频率安全稳定分析与控制.
基金资助:
国家自然科学基金(52077058);国网安徽电力有限公司科技项目(B31200220005)

A Prediction Method of New Power System Frequency Characteristics Based on Convolutional Neural Network

LU Wen’an¹, ZHU Qingxiao¹, LI Zhaowei², LIU Hui³, YU Yiping¹()

1. College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China
2. NARI Group Corporation/State Grid Electric Power Research Institute, Nanjing 211106, China
3. State Grid Anhui Electric Power Co., Ltd., Hefei 230022, China

Received:2023-03-03 Revised:2023-05-08 Accepted:2023-05-26 Online:2024-10-28 Published:2024-11-01

摘要/Abstract

摘要：

为了解决利用传统频率分析方法分析新能源高占比电网频率时存在计算量大、建模困难、计算速度与计算精度矛盾突出等问题,提出一种基于卷积神经网络(CNN)的新型电力系统频率特性预测方法.首先,利用一维CNN对新能源高占比电力系统在功率扰动下的主要频率指标进行预测,包括初始频率变化率、频率极值以及频率稳态值;并通过设置合理的输入特征以及对神经网络各参数的优化调整,提高了预测精度.在此基础上,进一步考虑扰动位置以及扰动类型的影响,利用数据降维的方法建立包含扰动信息的电力系统特征数据集,借鉴三原色通道原理构建输入特征,并利用扩展的二维CNN预测频率安全指标提高CNN在高占比新能源电网频率分析中的适应性.最后,在改进的BPA 10机39节点模型中进行算例验证,并与循环神经网络预测结果进行对比,结果表明所提方法具有较高的准确度和适应性.

关键词: 新型电力系统, 卷积神经网络, 频率安全指标, 扰动信息, 频率特性预测

Abstract:

In order to solve the problems existing in the traditional frequency analysis method for the frequency analysis of grids with a high proportion of new energy, such as the large amount of calculation, the difficulty of modeling, and the prominent contradiction between the calculation speed and the calculation accuracy, this paper proposes a new frequency characteristic prediction method for the new power system based on convolutional neural network (CNN). First, the main frequency indexes of the power system with a high proportion of new energy under power disturbances are predicted using one-dimensional CNN, including the initial frequency change rate, frequency extremum, and frequency steady-state value. The prediction accuracy is improved by setting reasonable input characteristics and optimizing the parameters of the neural network. Then, the impact of disturbance location and disturbance type is further considered, and the power system characteristic data set containing disturbance information is established by the method of data dimensionality reduction. The input characteristics are constructed by using the principle of three primary channels for reference, and the extended two-dimensional CNN is used to predict the frequency security index, which improves the adaptability of CNN in the frequency analysis of grids with a high proportion of new energy. Finally, the method is verified by an example in the improved BPA 10-machine 39-node model, and the results are compared with the prediction results of the recurrent neural network, which proves that the proposed method has a high accuracy and adaptability.

Key words: new power system, convolutional neural network (CNN), frequency safety index, disturbance information, frequency characteristic prediction

中图分类号:

TM716

陆文安, 朱清晓, 李兆伟, 刘辉, 余一平. 基于卷积神经网络的新型电力系统频率特性预测方法[J]. 上海交通大学学报, 2024, 58(10): 1500-1512.

LU Wen’an, ZHU Qingxiao, LI Zhaowei, LIU Hui, YU Yiping. A Prediction Method of New Power System Frequency Characteristics Based on Convolutional Neural Network[J]. Journal of Shanghai Jiao Tong University, 2024, 58(10): 1500-1512.

图/表 20

图1

图2

表1

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

表2

图14

图15

图16

表3

表4

不同神经网络模型预测结果对比

网络类型	$ε - M A E$	R²
RNN	0.0423	0.9024
一维CNN	0.0267	0.9400
二维CNN	0.0017	0.9684

表4

参考文献 24

[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]	PHURAILATPAM C, RATHER Z H, BAHRANI B, et al. Measurement-based estimation of inertia in AC microgrids[J]. IEEE Transactions on Sustainable Energy, 2020, 11(3): 1975-1984.
[3]	孟建辉, 彭嘉琳, 王毅, 等. 多约束下光储系统的灵活虚拟惯性控制方法[J]. 电工技术学报, 2019, 34(14): 3046-3058.
	MENG Jianhui, PENG Jialin, WANG Yi, et al. Multi-constrained flexible virtual inertial control method for photovoltaic energy storage system[J]. Transactions of China Electrotechnical Society, 2019, 34(14): 3046-3058.
[4]	王博, 杨德友, 蔡国伟. 高比例新能源接入下电力系统惯量相关问题研究综述[J]. 电网技术, 2020, 44(8): 2998-3007.
	WANG Bo, YANG Deyou, CAI Guowei. Review of research on power system inertia related issues in the context of high penetration of renewable power generation[J]. Power System Technology, 2020, 44(8): 2998-3007.
[5]	PEKER M, KOCAMAN A S, KARA B Y. Benefits of transmission switching and energy storage in power systems with high renewable energy penetration[J]. Applied Energy, 2018, 228: 1182-1197.
[6]	李元臣, 文云峰, 叶希, 等. 基于多新息辨识的电力系统节点惯量估计方法[J]. 电力自动化设备, 2022, 42(8): 89-95.
	LI Yuanchen, WEN Yunfeng, YE Xi, et al. Estimation method of power system nodal inertia based on multi-innovation identification[J]. Electric Power Automation Equipment, 2022, 42(8): 89-95.
[7]	CAO Y J, ZHANG Y, ZHANG H X, et al. Probabilistic optimal PV capacity planning for wind farm expansion based on NASA data[J]. IEEE Transactions on Sustainable Energy, 2017, 8(3): 1291-1300.
[8]	吕凯, 唐昊, 王珂, 等. 计及源荷双侧不确定性的跨区互联电网源网荷协同调度[J]. 电力系统自动化, 2019, 43(22): 38-45.
	LYU Kai, TANG Hao, WANG Ke, et al. Coordinated dispatching of source-grid-load for inter-regional power grid considering uncertainties of both source and load sides[J]. Automation of Electric Power Systems, 2019, 43(22): 38-45.
[9]	张子扬, 张宁, 杜尔顺, 等. 双高电力系统频率安全问题评述及其应对措施[J]. 中国电机工程学报, 2022, 42(1): 1-25.
	ZHANG Ziyang, ZHANG Ning, DU Ershun, et al. Review and countermeasures on frequency security issues of power systems with high shares of renewables and power electronics[J]. Proceedings of the CSEE, 2022, 42(1): 1-25.
[10]	文云峰, 杨伟峰, 林晓煌. 低惯量电力系统频率稳定分析与控制研究综述及展望[J]. 电力自动化设备, 2020, 40(9): 211-222.
	WEN Yunfeng, YANG Weifeng, LIN Xiaohuang. Review and prospect of frequency stability analysis and control of low-inertia power systems[J]. Electric Power Automation Equipment, 2020, 40(9): 211-222.
[11]	张怡. 基于深度学习的电力系统扰动后频率预测[D]. 济南: 山东大学, 2018.
	ZHANG Yi. Power system frequency prediction after disturbances based on deep learning[D]. Jinan: Shandong University, 2018.
[12]	杨挺, 赵黎媛, 王成山. 人工智能在电力系统及综合能源系统中的应用综述[J]. 电力系统自动化, 2019, 43(1): 2-14.
	YANG Ting, ZHAO Liyuan, WANG Chengshan. Review on application of artificial intelligence in power system and integrated energy system[J]. Automation of Electric Power Systems, 2019, 43(1): 2-14.
[13]	WEN S L, ZHAO T Y, WANG Y, et al. A deep learning method for power fluctuation identification from frequency fluctuations[C]// 2019 IEEE Power & Energy Society General Meeting. Atlanta, USA: IEEE, 2019: 1-5.
[14]	CHEN X Y, CHEN Z, HAN X Y, et al. A method of frequency features prediction of post-disturbance power system based on XGBoost algorithm[C]// 2020 IEEE/IAS Industrial and Commercial Power System Asia (I&CPS Asia). Weihai, China: IEEE, 2020: 1535-1540.
[15]	MA C H, WANG L, GAI C H, et al. Frequency security assessment for receiving-end system based on deep learning method[C]// 2020 IEEE/IAS Industrial and Commercial Power System Asia (I&CPS Asia). Weihai, China: IEEE, 2020: 831-836.
[16]	薄其滨, 王晓茹, 刘克天. 基于v-SVR的电力系统扰动后最低频率预测[J]. 电力自动化设备, 2015, 35(7): 83-88.
	BO Qibin, WANG Xiaoru, LIU Ketian. Minimum frequency prediction based on v-SVR for post-disturbance power system[J]. Electric Power Automation Equipment, 2015, 35(7): 83-88.
[17]	LONG X L, GUO J, HAO R, et al. Optical neural networks of handwriting recognition using optical scattering unit system[C]// Asia Communications and Photonics Conference/International Conference on Information Photonics and Optical Communications 2020. Beijing, China: Optica Publishing Group, 2020: 1-3.
[18]	马世龙, 乌尼日其其格, 李小平. 大数据与深度学习综述[J]. 智能系统学报, 2016, 11(6): 728-742.
	MA Shilong, WUNIRI Qiqige, LI Xiaoping. Deep learning with big data: State of the art and development[J]. CAAI Transactions on Intelligent Systems, 2016, 11(6): 728-742.
[19]	张泽超. 深度学习网络分布式训练方案研究与性能优化[D]. 杭州: 浙江大学, 2021.
	ZHANG Zechao. Optimization of distributed training strategies for deep learning networks[D]. Hangzhou: Zhejiang University, 2021.
[20]	ANTIOQUIA A M C, TAN D S, AZCARRAGA A, et al. ZipNet:ZFNet-level accuracy with 48 × fewer parameters[C]// 2018 IEEE Visual Communications and Image Processing. Taichung, China: IEEE, 2018: 1-4.
[21]	JILANI U, AKRAM N, ABBASI M, et al. Machine learning based leaves classifier using CNN and reduced VGG net model[C]// 2022 Global Conference on Wireless and Optical Technologies. Malaga, Spain:IEEE, 2022: 1-7.
[22]	林进钿. 基于深度学习的电力系统扰动后动态频率特征预测[D]. 成都: 西南交通大学, 2019.
	LIN Jintian. Power system post-disturbance dynamic frequency feature prediction based on deep learning[D]. Chengdu: Southwest Jiaotong University, 2019.
[23]	郑超, 王士元, 张波琦, 等. 光伏高渗透电网动态频率特性及应对措施[J]. 电网技术, 2019, 43(11): 4064-4073.
	ZHENG Chao, WANG Shiyuan, ZHANG Boqi, et al. Dynamic frequency characteristics and countermeasures of power grid with highly penetrated photovoltaics[J]. Power System Technology, 2019, 43(11): 4064-4073.
[24]	陈俊佟. t-SNE结合支持向量机的降维分类研究[D]. 大连: 大连理工大学, 2021.
	CHEN Juntong. Research on dimensionality reduction classification of t-SNE combined with support vector machine[D]. Dalian: Dalian University of Technology, 2021.

编号	电力系统运行特征量
1	初始时刻发电机节点输出电磁功率
2	初始时刻发电机节点机械功率
3	初始时刻各节点电压幅值
4	初始时刻各节点电压相角
5	初始时刻各节点有功负荷
6	初始时刻新能源出力
7	故障瞬间发电机节点输出电磁功率
8	故障瞬间发电机节点机械功率
9	故障瞬间各节点电压幅值
10	故障瞬间各节点电压相角
11	故障瞬间各节点有功负荷
12	故障瞬间新能源节点新能源出力
13	发电机节点有功备用容量
14	发电机节点惯性时间常数

数据集批次	训练集		验证集
数据集批次	损失函数	R²	损失函数	R²
1	0.0014	0.9747	0.0012	0.9764
2	0.0013	0.9753	0.0014	0.9710
3	0.0013	0.9758	0.0017	0.9674
4	0.0015	0.9718	0.0018	0.9539
5	0.0014	0.9726	0.0013	0.9731

基于卷积神经网络的新型电力系统频率特性预测方法

A Prediction Method of New Power System Frequency Characteristics Based on Convolutional Neural Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 20

参考文献 24

相关文章 15

编辑推荐

Metrics

本文评价

预测指标	R²	ε_MAE
频率变化率	0.94	0.03
频率极值	0.96	0.03
频率稳态值	0.93	0.02

[1]	邓小宇, 刘牧阳, 常喜强, 南东亮, 莫若, 陈俊儒. 新型电力系统点对网惯性支撑能力在线监测方法[J]. 上海交通大学学报, 2024, 58(9): 1390-1399.
[2]	于淼, 胡敬轩, 张寿志, 魏静静, 孙建群, 吴屹潇. 基于PMU梯度动态偏差的新型电力系统快速稳定性[J]. 上海交通大学学报, 2024, 58(1): 40-49.
[3]	詹可, 朱仁传. 一种CNN-LSTM船舶运动极值预报模型[J]. 上海交通大学学报, 2023, 57(8): 963-971.
[4]	李擎, 皇甫玉彬, 李江昀, 杨志方, 陈鹏, 王子涵. UConvTrans:全局和局部信息交互的双分支心脏图像分割[J]. 上海交通大学学报, 2023, 57(5): 570-581.
[5]	唐震, 郝丽花, 冯静. 局部频率测量数据驱动的电力系统功率缺额在线评估方法[J]. 上海交通大学学报, 2023, 57(4): 403-411.
[6]	万安平, 杨洁, 缪徐, 陈挺, 左强, 李客. 基于注意力机制与神经网络的热电联产锅炉负荷预测[J]. 上海交通大学学报, 2023, 57(3): 316-325.
[7]	王者蓝, 赵宏杰, 赵凡, 沈晨晨, 吴佳伟. 基于卷积神经网络与滤波融合算法的某惯导系统剩余寿命预测模型建立[J]. 空天防御, 2023, 6(1): 70-77.
[8]	. 基于锥型体素建模和单目相机的鸟瞰图语义分割和体素语义分割[J]. J Shanghai Jiaotong Univ Sci, 2023, 28(1): 100-113.
[9]	曾国治, 魏子清, 岳宝, 丁云霄, 郑春元, 翟晓强. 基于CNN-RNN组合模型的办公建筑能耗预测[J]. 上海交通大学学报, 2022, 56(9): 1256-1261.
[10]	全大英, 陈赟, 唐泽雨, 李世通, 汪晓锋, 金小萍. 基于双通道卷积神经网络的雷达信号识别[J]. 上海交通大学学报, 2022, 56(7): 877-885.
[11]	吴庶宸, 戚宗锋, 李建勋. 基于深度学习的智能全局灵敏度分析[J]. 上海交通大学学报, 2022, 56(7): 840-849.
[12]	赵勇, 苏丹. 基于4种长短时记忆神经网络组合模型的畸形波预报[J]. 上海交通大学学报, 2022, 56(4): 516-522.
[13]	张硕, 李薇, 李英姿, 刘强, 曾鸣. 面向新型电力系统的可再生能源绿色电力证书差异化配置模型[J]. 上海交通大学学报, 2022, 56(12): 1561-1571.
[14]	陶海红, 闫莹菲. 一种基于GA-CNN的网络化雷达节点遴选算法[J]. 空天防御, 2022, 5(1): 1-5.
[15]	武光利, 郭振洲, 李雷霆, 王成祥. 融合FCN和LSTM的视频异常事件检测[J]. 上海交通大学学报, 2021, 55(5): 607-614.