METHOD FOR MISSING DATA IMPUTATION IN DISTRIBUTION NETWORK BASED ON GRADIENT PENALTY GENERATIVE ADVERSARIAL NETWORK

Lyu Pengpeng; Bu Qiangsheng; Guo Ye; Luo Fei

doi:10.19912/j.0254-0096.tynxb.2024-0160

PDF(1546 KB)

Acta Energiae Solaris Sinica ›› 2025, Vol. 46 ›› Issue (5) : 185-192. DOI: 10.19912/j.0254-0096.tynxb.2024-0160

METHOD FOR MISSING DATA IMPUTATION IN DISTRIBUTION NETWORK BASED ON GRADIENT PENALTY GENERATIVE ADVERSARIAL NETWORK

Lyu Pengpeng, Bu Qiangsheng, Guo Ye, Luo Fei

Author information +

History +

Abstract

With the rapid development of renewable energy such as wind power and photovoltaic, the measurement data of distribution networks present characteristics such as non-linearity, high-dimensionality, and time-variation, significantly increasing the difficulty of data imputation. This paper proposes a method for missing data imputation in distribution networks based on the gradient penalty generative adversarial network. The structure and parameters of the generator and discriminator are designed according to the characteristics of the distribution network data, and a gradient penalty term is introduced in the training of the generative adversarial network to improve convergence performance. This method is data-driven, understanding high-dimensional, non-linear features in the data that are difficult to characterize through unsupervised learning and does not require modeling of the specific topology of the distribution network, enhancing the applicability of the data restoration method. The results of the example show that this method has higher imputation accuracy compared to traditional generative adversarial networks. When the proportion of missing data is 10%, the accuracy of data repair is increased by 18.9%.

Key words

power measurement / big data / generative adversarial network / unsupervised learning / missing data imputation

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

Lyu Pengpeng, Bu Qiangsheng, Guo Ye, Luo Fei. METHOD FOR MISSING DATA IMPUTATION IN DISTRIBUTION NETWORK BASED ON GRADIENT PENALTY GENERATIVE ADVERSARIAL NETWORK[J]. Acta Energiae Solaris Sinica. 2025, 46(5): 185-192 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0160

References

[1] 王赞, 陈光, 董晓, 等. 基于工业互联网的智慧能源服务系统架构研究[J]. 电力系统保护与控制, 2020, 48(3): 77-83.
WANG Z, CHEN G, DONG X, et al.Research on the architecture of smart energy service system based on industrial Internet[J]. Power system protection and control, 2020, 48(3): 77-83.
[2] 胡守超. 智能电网数字孪生技术发展方向及应用[J]. 太阳能学报, 2023, 44(11): 576.
HU S C.Development direction and application of digital twinning technology in smart grid[J]. Acta energiae solaris sinica, 2023, 44(11): 576.
[3] 刘洪波, 刘珅诚, 盖雪扬, 等. 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.
LIU H B, LIU S C, GAI X Y, et al.Overview of active distribution network planning with high proportion of new energy access[J]. Power generation technology, 2024, 45(1): 151-161.
[4] 王翔宇, 陈武晖, 郭小龙, 等. 发电系统数字化研究综述[J]. 发电技术, 2024, 45(1): 120-141.
WANG X Y, CHEN W H, GUO X L, et al.Review of research on the digitalization of power generation system[J]. Power generation technology, 2024, 45(1): 120-141.
[5] 李富盛, 陈伟松, 钱斌, 等. 面向低压配电网智能电表误差监测的LightGBM-EM-EC多变量缺失数据高效重建[J]. 中国电机工程学报, 2022, 42(S1): 95-105.
LI F S, CHEN W S, QIAN B, et al.Efficient reconstruction of multivariate missing data for smart meter error monitoring in low voltage distribution network based on LightGBM-EM-EC[J]. Proceedings of the CSEE, 2022, 42(S1): 95-105.
[6] 沈琳, 陈千红, 谭红专. 缺失数据的识别与处理[J]. 中南大学学报(医学版), 2013, 38(12): 1289-1294.
SHEN L, CHEN Q H, TAN H Z.Identification and treatment of missing data[J]. Journal of Central South University (medical science), 2013, 38(12): 1289-1294.
[7] 陈婧, 林超, 薛迎卫, 等. 智能电网电量异常数据的识别和修复研究[J]. 自动化仪表, 2023, 44(10): 80-84.
CHEN J, LIN C, XUE Y W, et al.Research on identification and repair of power abnormal data in smart grid[J]. Process automation instrumentation, 2023, 44(10): 80-84.
[8] 王晓虎. 张量分解理论下的10 kV配电网电压数据修复方法[J]. 电力系统及其自动化学报, 2022, 34(7): 148-152.
WANG X H.Voltage data repair method for 10 kV distribution network under tensor decomposition theory[J]. Proceedings of the CSU-EPSA, 2022, 34(7): 148-152.
[9] 孙川. 基于时空和特征相关性的风电缺失数据修复研究[D]. 武汉: 华中科技大学, 2022.
SUN C.Research on restoration of wind power missing data based on temporal and spatial correlation and feature correlation[D]. Wuhan: Huazhong University of Science and Technology, 2022.
[10] MIAO X Y, WU Y Y, CHEN L, et al.An experimental survey of missing data imputation algorithms[J]. IEEE transactions on knowledge and data engineering, 2023, 35(7): 6630-6650.
[11] 刘德澳, 董明刚, 叶威, 等. 基于图神经网络的非均匀风电场功率预测方法[J]. 太阳能学报, 2023, 44(11): 279-286.
LIU D A, DONG M G, YE W, et al.Power prediction method of non-uniform wind farm layout based on graph neural network[J]. Acta energiae solaris sinica, 2023, 44(11): 279-286.
[12] 董文康, 吴雨芯, 姚琦, 等. 基于深度强化学习的海上风电机组状态维护与备件库存联合优化[J]. 太阳能学报, 2023, 44(12): 190-199.
DONG W K, WU Y X, YAO Q, et al.Joint optimization of state maintenance and spare parts inventory of offshore wind turbines based on deep reinforcement learning[J]. Acta energiae solaris sinica, 2023, 44(12): 190-199.
[13] 王守相, 陈海文, 潘志新, 等. 采用改进生成式对抗网络的电力系统量测缺失数据重建方法[J]. 中国电机工程学报, 2019, 39(1): 56-64, 320.
WANG S X, CHEN H W, PAN Z X, et al.A reconstruction method for missing data in power system measurement using an improved generative adversarial network[J]. Proceedings of the CSEE, 2019, 39(1): 56-64, 320.
[14] 余磊, 贾科, 温志文, 等. 计及量测数据丢失的主动配电网电流保护自适应整定方法[J]. 电力系统自动化, 2022, 46(15): 145-152.
YU L, JIA K, WEN Z W, et al.Adaptive setting method of over current protection in active distribution network considering measurement data missing[J]. Automation of electric power systems, 2022, 46(15): 145-152.
[15] 杨玉莲, 齐林海, 王红, 等. 基于生成对抗和双重语义感知的配电网量测数据缺失重构[J]. 电力系统自动化, 2020, 44(18): 46-54.
YANG Y L, QI L H, WANG H, et al.Reconstruction of missing measurement data in distribution network based on generative adversarial network and double semantic perception[J]. Automation of electric power systems, 2020, 44(18): 46-54.
[16] ZHANG W J, LUO Y H, ZHANG Y, et al.SolarGAN: multivariate solar data imputation using generative adversarial network[J]. IEEE transactions on sustainable energy, 2021, 12(1): 743-746.
[17] 皇甫汉聪. 基于次序依赖的电力数据缺失自动修复方法研究[J]. 自动化与仪器仪表, 2020(12): 233-236.
HUANGPU H C.Research on automatic repair method of power data missing based on order dependence[J]. Automation & instrumentation, 2020(12): 233-236.
[18] CRESWELL A, WHITE T, DUMOULIN V, et al.Generative adversarial networks: an overview[J]. IEEE signal processing magazine, 35(1): 53-65.
[19] LI C Y, LIU X X, LI S T.Transformer meets GAN: cloud-free multispectral image reconstruction via multisensor data fusion in satellite images[J]. IEEE transactions on geoscience and remote sensing, 2023, 61: 5408513.
[20] WU Y Y, WANG J, MIAO X Y, et al.Differentiable and scalable generative adversarial models for data imputation[J]. IEEE transactions on knowledge and data engineering, 2024, 36(2): 490-503.
[21] GULRAJANI I, AHMED F, ARJOVSKY M, et al.Improved training of Wasserstein GANs[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, California, USA, 2017: 5769-5779.
[22] NING K Q, ZHANG Z H, HAN K, et al.Multi-frame su-per-resolution algorithm based on a WGAN[J]. IEEE access, 2021, 9: 85839-85851.
[23] GAO X, DENG F, YUE X H.Data augmentation in fault diagnosis based on the Wasserstein generative adversarial network with gradient penalty[J]. Neurocomputing, 2020, 396: 487-494.