应用深度学习网络的变电站二次测量回路误差评估

doi:10.11930/j.issn.1004-9649.202406003

摘要/Abstract

摘要：

变电站的测量及保护装置伴随环境以及自身使用磨损易引起其监测的电流变动产生误差，从而导致测量回路面临误动拒动风险，常规监控方法难以检测该类幅度变化，基于此提出一种条件生成对抗网络（conditional generative adversarial nets，CGAN）和改进的长短期记忆（long short-term memory，LSTM）网络的误差评估方法。首先，采集正常运行下测量回路的电流数据，引入CGAN方法进行误差数据的增强生成；其次，对生成后的数据进行经验模态分解（empirical mode decomposition，EMD）构成样本并选择最优特征集；为进一步评估误差状态，采用改进的长短期记忆（long short-term memory，LSTM）算法训练模型；最后，搭建PSCAD/EMTDC仿真模型验证本文所提方法的可靠性和准确性。测试实验结果表明：本文所采用的新方法能够可靠地评估二次系统测量回路2%的误差状态。

关键词: 测量回路, 生成式对抗网络, 误差评估

Abstract:

Measuring and protection devices in substations are susceptible to errors in the monitored current variations due to environmental and wear and tear, which leads to the risk of inadvertent refusal of the measuring circuits, and it is difficult to detect such amplitude variations by the conventional monitoring methods, based on which, this paper proposes a conditional generative adversarial network (CGAN) and an improved long and short-term memory network (STMN) for the error assessment method. Firstly, the current data of the measurement loop under normal operation is obtained, and the CGAN method is introduced to enhance the generation of error data; secondly, the EMD decomposition of the generated data is performed to form samples and the optimal set of features is selected; in order to further evaluate the error state, the improved LSTM algorithm is used to train the model; finally, a PSCAD/EMTDC simulation model is constructed to verify the reliability and accuracy of the methodology presented in this paper. Finally, a PSCAD/EMTDC simulation model is built to verify the reliability and accuracy of the proposed method. Finally, a PSCAD/EMTDC simulation model is constructed to verify the reliability and accuracy of the proposed method. The test results show that the new method adopted in this paper can reliably evaluate the error state of 2% of the measurement loop of the secondary system.

Key words: measurement loops, generative adversarial networks, error assessment

吴江雄, 刘千宽, 阳国燕, 蒋连钿. 应用深度学习网络的变电站二次测量回路误差评估[J]. 中国电力, 2025, 58(6): 206-212.

WU Jiangxiong, LIU Qiankuan, YANG Guoyan, JIANG Liandian. Secondary Measurement Loop Error Assessment in Substations with Application of Deep Learning Networks[J]. Electric Power, 2025, 58(6): 206-212.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202406003

https://www.electricpower.com.cn/CN/Y2025/V58/I6/206

图/表 10

图 1 变电站二次测量回路

Fig.1 Substation secondary measurement circuits

图 2 CGAN 结构

Fig.2 Structure of CGAN

图 3 CGAN 结构

Fig.3 Structure of CGAN

图 4 二次系统测量回路误差评估模型

Fig.4 Secondary system measurement loop error assessment model

图 5 生成器和判别器的损失函数曲线

Fig.5 Loss function curves of generator and discriminator

图 6 通过CGAN模型生成的数据

Fig.6 Data generated by the CGAN model

表 1 厂变电流的误差状态对应的标签和编码

Table 1 The label and code corresponding to the error state of the transformer current

误差状态	标签	编码
–2%~2%	1	[0 0 1]
≥2%	2	[0 1 0]
≤–2%	3	[1 0 0]

图 7 四种网络准确率随迭代次数变化情况

Fig.7 The accuracy of the four networks varies with the number of iterations

图 8 四种网络损失函数随迭代次数变化情况

Fig.8 The four network loss functions vary with the number of iterations

图 9 数据生成前后准确率对比

Fig.9 Accuracy comparison before and after data generation

参考文献 31

1	计荣荣, 王梦芝, 周国伟, 等. 基于云端优化的智能变电站二次系统快速测试建模与系统设计[J]. 中国电力, 2025, 58 (5): 152- 157. DOI
	JI Rongrong, WANG Mengzhi, ZHOU Guowei, et al. Cloud-based optimized intelligent substation secondary system rapid test model and system design[J]. Electric Power, 2025, 58 (5): 152- 157. DOI
2	金国锋, 杨世峰, 刘玲玲, 等. 基于二维混沌映射正余弦算法的智能变电站虚回路自动连接技术[J]. 中国电力, 2024, 57 (8): 152- 158.
	JIN Guofeng, YANG Shifeng, LIU Lingling, et al. Automatic connection of virtual circuits in smart substations based on chaotic two-dimensional mapping sine and cosine algorithm[J]. Electric Power, 2024, 57 (8): 152- 158.
3	曹一家, 赵一睿, 李勇, 等. 考虑针对变电站网络攻击风险的脆弱元件筛选方法[J]. 电力系统自动化, 2023, 47 (8): 25- 33. DOI
	CAO Yijia, ZHAO Yirui, LI Yong, et al. Vulnerable component screening method considering cyber-attack risk in substation[J]. Automation of Electric Power Systems, 2023, 47 (8): 25- 33. DOI
4	严亚兵, 褚旭, 肖豪龙, 等. 基于改进支持向量机的数字化变电站安全措施生成技术[J]. 中国电力, 2023, 56 (10): 194- 201.
	YAN Yabing, CHU Xu, XIAO Haolong, et al. Automatic generation technology of safety measures for digital substation based on improved support vector machine[J]. Electric Power, 2023, 56 (10): 194- 201.
5	刘健, 朱青松, 王毅钊, 等. 基于变电站稳态监测信息的小电阻接地系统单相接地检测性能提升方法[J]. 浙江电力, 2023, 42 (4): 1- 8.
	LIU Jian, ZHU Qingsong, WANG Yizhao, et al. Improvement of single-phase grounding fault detection for small resistance grounding systems based on steady-state monitoring information in the substation[J]. Zhejiang Electric Power, 2023, 42 (4): 1- 8.
6	朱智, 张永宏, 吕帅, 等. 基于数据通信建模技术的移动式智能变电站安措防误预警系统研究与应用[J]. 内蒙古电力技术, 2023, 41 (4): 54- 59.
	ZHU Zhi, ZHANG Yonghong, LYU Shuai, et al. Research and application on security precaution misoperation early warning system of mobile intelligent substation based on data communication modeling technology[J]. Inner Mongolia Electric Power, 2023, 41 (4): 54- 59.
7	李振华, 郑严钢, 李振兴, 等. 基于传递熵和小波神经网络的电子式电压互感器误差预测[J]. 电测与仪表, 2021, 58 (3): 146- 152.
	LI Zhenhua, ZHENG Yangang, LI Zhenxing, et al. Error prediction of electronic voltage transformer based on transfer entropy and wavelet neural network[J]. Electrical Measurement & Instrumentation, 2021, 58 (3): 146- 152.
8	赵鹏, 马克琪, 李红斌, 等. 计及一次电压波动的电压互感器误差状态在线评估方法[J]. 高电压技术, 2022, 48 (3): 1042- 1051.
	ZHAO Peng, MA Keqi, LI Hongbin, et al. On-line error evaluation method of instrument voltage transformers considering primary voltage fluctuations[J]. High Voltage Engineering, 2022, 48 (3): 1042- 1051.
9	高书垚, 安泰, 宋剑. 基于熵权-正态云模型的智能电能表状态评估研究[J]. 电测与仪表, 2022, 59 (1): 190- 194.
	GAO Shuyao, AN Tai, SONG Jian. Status evaluation of smart meter based on entropy weight-normal cloud model[J]. Electrical Measurement & Instrumentation, 2022, 59 (1): 190- 194.
10	南东亮, 王维庆, 张陵, 等. 基于关联规则挖掘与组合赋权-云模型的电网二次设备运行状态风险评估[J]. 电力系统保护与控制, 2021, 49 (10): 67- 76.
	NAN Dongliang, WANG Weiqing, ZHANG Ling, et al. Risk assessment of the operation state of power grid secondary equipment based on association rule mining and combination weighting-cloud model[J]. Power System Protection and Control, 2021, 49 (10): 67- 76.
11	顾佳浩, 淡淑恒. 考虑HI理论和在线监测误差的配电网可靠性评估[J]. 电力系统及其自动化学报, 2021, 33 (4): 127- 134.
	GU Jiahao, DAN Shuheng. Reliability evaluation on distribution network considering HI theory and online monitoring error[J]. Proceedings of the CSU-EPSA, 2021, 33 (4): 127- 134.
12	陈海涛, 杨军, 施迎春, 等. 基于云模型与马尔科夫链的继电保护装置寿命预测方法[J]. 电力系统保护与控制, 2019, 47 (16): 94- 100.
	CHEN Haitao, YANG Jun, SHI Yingchun, et al. Life prediction method of relay protection device based on could model and Markov Chain[J]. Power System Protection and Control, 2019, 47 (16): 94- 100.
13	李俊卿, 李斯璇, 陈雅婷, 等. 一种基于CGAN-CNN的同步电机转子绕组匝间短路故障诊断方法[J]. 电力自动化设备, 2021, 41 (8): 169- 174.
	LI Junqing, LI Sixuan, CHEN Yating, et al. Fault diagnosis method of inter-turn short circuit of rotor winding of synchronous motor based on CGAN-CNN[J]. Electric Power Automation Equipment, 2021, 41 (8): 169- 174.
14	孙宁, 陈争, 侯小强, 等. 基于MGP-GAN的27.5 kV电缆故障诊断方法[J]. 电气自动化, 2023, 45 (4): 88- 90.
	SUN Ning, CHEN Zheng, HOU Xiaoqiang, et al. Fault diagnosis method of 27.5 kV cable based on MGP-GAN[J]. Electrical Automation, 2023, 45 (4): 88- 90.
15	邵振国, 张承圣, 陈飞雄, 等. 生成对抗网络及其在电力系统中的应用综述[J]. 中国电机工程学报, 2023, 43 (3): 987- 1004.
	SHAO Zhenguo, ZHANG Chengsheng, CHEN Feixiong, et al. A review on generative adversarial networks for power system applications[J]. Proceedings of the CSEE, 2023, 43 (3): 987- 1004.
16	江善和, 李伟, 徐小艳, 等. 基于变分模态分解改进生成对抗网络的短期风电功率预测[J]. 综合智慧能源, 2024, 46 (2): 28- 35. DOI
	JIANG Shanhe, LI Wei, XU Xiaoyan, et al. Short-term wind power forecasting based on variational mode decomposition and generative adversarial networks[J]. Integrated Intelligent Energy, 2024, 46 (2): 28- 35. DOI
17	黄海荣, 吴君, 郁丹, 等. 基于经验模态条件生成对抗网络的短期负荷预测[J]. 自动化仪表, 2022, 43 (12): 25- 29, 37.
	HUANG Hairong, WU Jun, YU Dan, et al. Short-term load forecasting based on empirical modal conditional generative adversarial network[J]. Process Automation Instrumentation, 2022, 43 (12): 25- 29, 37.
18	邵晨颖, 刘友波, 邵安海, 等. 基于生成对抗网络与局部电流相量的配电网拓扑鲁棒辨识[J]. 电力系统自动化, 2023, 47 (1): 55- 62. DOI
	SHAO Chenying, LIU Youbo, SHAO Anhai, et al. Robust identification for distribution network topology based on generative adversarial network and partial current phasor[J]. Automation of Electric Power Systems, 2023, 47 (1): 55- 62. DOI
19	罗萍萍, 盛奥, 林济铿, 等. 基于CGAN台风气象下负荷场景生成[J]. 中国电力, 2025, 58 (2): 176- 185. DOI
	LUO Pingping, SHENG Ao, LIN Jikeng, et al. CGAN-based load scenario generation under typhoon weather[J]. Electric Power, 2025, 58 (2): 176- 185. DOI
20	孙浩, 万灿, 曹照静, 等. 基于条件生成对抗网络曲线生成的短期负荷概率预测[J]. 电力系统自动化, 2023, 47 (23): 189- 199. DOI
	SUN Hao, WAN Can, CAO Zhaojing, et al. Short-term load probabilistic forecasting based on conditional generative adversarial network curve generation[J]. Automation of Electric Power Systems, 2023, 47 (23): 189- 199. DOI
21	杨再丞, 孙勇. 基于空间简化和CNN-LSTM的区域风电功率日前网格预测方法[J]. 东北电力大学学报, 2024, 44 (2): 35- 41.
	YANG Zaicheng, SUN Yong. Foreahead grid prediction of regional wind power basedon spatial reduction and CNN-LSTM[J]. Journal of Northeast Electric Power University, 2024, 44 (2): 35- 41.
22	张哲, 王勃. 基于CBAM-LSTM的风电集群功率短期预测方法[J]. 东北电力大学学报, 2024, 44 (1): 1- 8.
	ZHANG Zhe, WANG Bo. Short-term power prediction method of wind power cluster based on CBAM-LSTM[J]. Journal of Northeast Electric Power University, 2024, 44 (1): 1- 8.
23	袁俊球, 王迪, 谢小锋, 等. 基于广义天气分类的ICEEMDAN-LSTM网络光伏发电功率短期预测[J]. 综合智慧能源, 2024, 46 (9): 53- 60. DOI
	YUAN Junqiu, WANG Di, XIE Xiaofeng, et al. Study of short-term PV power prediction based on ICEEMDAN-LSTM networks under generalized weather classifications[J]. Integrated Intelligent Energy, 2024, 46 (9): 53- 60. DOI
24	薛飞, 李宏强, 李旭涛, 等. 基于LSTM神经网络的双馈风机控制参数辨识方法[J]. 中国电力, 2023, 56 (6): 31- 39.
	XUE Fei, LI Hongqiang, LI Xutao, et al. Identification method for control parameters of doubly-fed induction generator based on LSTM neural network[J]. Electric Power, 2023, 56 (6): 31- 39.
25	张宏杰, 陈贵凤, 闫宏伟, 等. 基于SMOTE与Bayes优化的LSTM网络变压器故障诊断[J]. 中国电力, 2023, 56 (10): 164- 170.
	ZHANG Hongjie, CHEN Guifeng, YAN Hongwei, et al. Fault diagnosis of LSTM network tansformer based on SMOTE and Bayes optimization[J]. Electric Power, 2023, 56 (10): 164- 170.
26	杨协伟, 方文田, 蔡伟杰. 一种提升金属氧化物避雷器阻性电流计算精度的方法[J]. 浙江电力, 2023, 42 (8): 92- 99.
	YANG Xiewei, FANG Wentian, CAI Weijie. A method of improving calculation accuracy of resistive current for metal oxide arrester[J]. Zhejiang Electric Power, 2023, 42 (8): 92- 99.
27	金恩淑, 陈和平, 夏国武. 基于开关函数的晶闸管整流器差动保护新方法[J]. 东北电力大学学报, 2023, 43 (2): 31- 38.
	JIN Enshu, CHEN Heping, XIA Guowu. A new differential protection method of thyristor converter in rectifier station based on wwitching function[J]. Journal of Northeast Electric Power University, 2023, 43 (2): 31- 38.
28	姚创, 魏菊芳, 于顺智, 等. 基于流敏电阻的电磁式电压互感器高压侧消谐技术研究[J]. 广东电力, 2023, 36 (12): 97- 105.
	YAO Chuang, WEI Jufang, YU Shunzhi, et al. Research on harmonic elimination technology at high voltage terminal of inductive voltage transformer based on current dependent resistor[J]. Guangdong Electric Power, 2023, 36 (12): 97- 105.
29	马文成. 电压互感器高压熔断器熔管更换辅助工具研制[J]. 内蒙古电力技术, 2023, 41 (2): 59- 63.
	MA Wencheng. Development of auxiliary tool for fuse tube replacement of high voltage fuse of voltage transformer[J]. Inner Mongolia Electric Power, 2023, 41 (2): 59- 63.
30	谷相宏, 郭子晗, 张程, 等. 正弦波调制全光纤电流互感器光回路故障预警方法[J]. 广东电力, 2023, 36 (7): 50- 59.
	GU Xianghong, GUO Zihan, ZHANG Cheng, et al. Fault early warning method for FOCT optical circuit of sine wave modulation[J]. Guangdong Electric Power, 2023, 36 (7): 50- 59.
31	乔胜亚, 周鸿铃, 李光茂, 等. 基于介损和电容量的电流互感器运行分布规律及现场应用研究[J]. 广东电力, 2023, 36 (3): 84- 91.
	QIAO Shengya, ZHOU Hongling, LI Guangmao, et al. Research on operation distribution law and field application of current transformer based on dielectric loss and capacitance[J]. Guangdong Electric Power, 2023, 36 (3): 84- 91.

[1]	尹相国, 刘慧娟, 张冠育, 毛玉荣, 龚世玉. 基于VAE-WSVM的二次采样回路测试校核评估[J]. 中国电力, 2025, 58(3): 162-167.
[2]	易亚文, 江传宾, 龚世玉, 秦小元, 赵静朴, 李振兴. 基于因子分析与统计学技术的保护测量回路误差评估[J]. 中国电力, 2023, 56(12): 183-190.
[3]	易亚文, 赵静朴, 柳灿, 蒋焮焮, 李方玖, 李振兴. 基于改进ED及CM-BPNN算法的保护测量回路误差状态评估方法[J]. 中国电力, 2023, 56(11): 143-150.