电力领域数据驱动建模实践与思考Practice and reflection on data-driven modeling in electric power domain
王慧芳,叶睿恺,罗斌,张波,吴雪峰,刘建敏
WANG Huifang,YE Ruikai,LUO Bin,ZHANG Bo,WU Xuefeng,LIU Jianmin
摘要(Abstract):
数据驱动建模综合利用理论、试验及数据三大研究范式且模型应用快速,将在解决新型电力系统技术问题中得到更多的应用。为此,对电力领域数据驱动建模的实践情况进行了总结并提出了相关思考。首先,介绍了电力领域数据驱动建模的技术与应用现状。然后,分析了针对理论建模性能欠佳问题的数据驱动建模实践情况并总结了建模过程中的共性步骤,介绍了基于电力文本数据和电力图像数据的数据驱动建模实践情况并总结了相关经验与体会。最后,对电力领域数据驱动建模的定义、条件与步骤、优势与风险等提出了一些理解与思考,讨论并总结了电力领域数据驱动建模应重视的若干问题。
Data-driven modeling,with its integrated use of such research paradigms as theory,experiment,and data,as well as its rapid application,will be more widely used for solving new problems in power system technology. To this end,the practice of data-driven modeling in the electric power domain is summarized and reflections concerning this matter are presented. First,data-driven modeling and its application in the electric power domain are presented. Then,the data-driven modeling practice for the poor performance of theoretical modeling is analyzed,and the common procedures in the modeling process are summarized. The practice of data-driven modeling based on power text data and power image data is introduced,and relevant experience and reflections are summarized. Finally,some comprehensions and reflections on the definition,conditions and procedures,advantages,and risks of data-driven modeling in the electric power domain are presented,and some issues on which importance ought to be attached in data-driven modeling in the electric power domain are discussed and summarized.
关键词(KeyWords):
电力领域;数据驱动建模;新型电力系统;建模条件;建模风险
electric power domain;data-driven modeling;new-type power system;modeling conditions;modeling risk
基金项目(Foundation): 国家自然科学基金联合基金资助项目(U2166204);; 国网浙江省电力有限公司双创资助项目(B711JZ21000P)
作者(Author):
王慧芳,叶睿恺,罗斌,张波,吴雪峰,刘建敏
WANG Huifang,YE Ruikai,LUO Bin,ZHANG Bo,WU Xuefeng,LIU Jianmin
DOI: 10.19585/j.zjdl.202210001
参考文献(References):
- [1]辛保安,郭铭群,王绍武,等.适应大规模新能源友好送出的直流输电技术与工程实践[J].电力系统自动化,2021,45(22):1-8.
- [2]董雪涛,冯长有,朱子民,等.面向以新能源为主体的新型电力系统仿真工具研究初探[J/OL].电力系统自动化:1-11.[2021-11-25].http://kns.cnki.net/kcms/detail/32.1180.TP.20211124.1503.006.html.
- [3] CHEN J Q,WU W C,ROALD L A.Data-driven piecewise linearization for distribution three-phase stochastic power flow[J].IEEE Transactions on Smart Grid,2022,13(2):1035-1048.
- [4]郑翔,王慧芳,姜宽,等.机理与数据融合驱动的含IIDG配电网短路电流计算方法[J].电力自动化设备,2021,41(1):41-48.
- [5]盛戈皞,钱勇,罗林根,等.面向新型电力系统的电力设备运行维护关键技术及其应用展望[J].高电压技术,2021,47(9):3072-3084.
- [6]刘梓权,王慧芳,曹靖,等.基于卷积神经网络的电力设备缺陷文本分类模型研究[J].电网技术,2018,42(2):644-651.
- [7]万吉林,王慧芳,管敏渊,等.基于Faster R-CNN和UNet的变电站指针式仪表读数自动识别方法[J].电网技术,2020,44(8):3097-3105.
- [8]刘迪一.AlphaFold即将解决生物学最大挑战之一[J].世界科学,2021(2):24-25.
- [9]谢小瑜,周俊煌,张勇军,等.基于W-BiLSTM的可再生能源超短期发电功率预测方法[J].电力系统自动化,2021,45(8):175-184.
- [10]杜一星,胡志坚,李犇,等.基于双向门控循环单元的电力系统暂态稳定评估[J].电力系统自动化,2021,45(20):103-112.
- [11]张晨宇,王慧芳,叶晓君.基于XGBoost算法的电力系统暂态稳定评估[J].电力自动化设备,2019,39(3):77-83.
- [12] WANG H F,ZHANG C Y,LIN D Y,HE B T.An artificial intelligence based method for evaluating power grid node importance using Network Embedding and Support Vector Regression[J]. Frontiers of Information Technology&Electronic Engineering,2019,20(6):816-828.
- [13]王慧芳,张晨宇.采用极限梯度提升算法的电力系统电压稳定裕度预测[J].浙江大学学报(工学版),2020,54(3):606-613.
- [14]郑一斌,王慧芳,张磊,等.基于LightGBM算法的配网单相接地故障区段定位方法[J].电力自动化设备,2021,41(12):54-61.
- [15]叶睿恺,王慧芳,张森,等.数据驱动的含IIDG配电网短路电流计算多输出模型[J].电力自动化设备,2022,42(9):119-125.
- [16]邵冠宇,王慧芳,何奔腾.电网设备缺陷文本的质量评价与提升方法[J].电网技术,2019,43(4):1472-1479.
- [17]郑翔,王慧芳,严娴峥,等.用于APP的缺陷文本自动分类与质量保证方法[J].电力系统及其自动化学报,2020,32(7):131-136.
- [18] WANG H F,LIU Z Q.An error recognition method for power equipment defect records based on knowledge graph technology[J].Frontiers of Information Technology&Electronic Engineering,2019,20(3):1564-1577.
- [19]曹靖,陈陆燊,邱剑,等.基于语义框架的电网缺陷文本挖掘技术及其应用[J].电网技术,2017,41(2):637-643.
- [20]刘梓权,王慧芳.基于知识图谱技术的电力设备缺陷记录检索方法[J].电力系统自动化,2018,42(14):158-164.
- [21]邵冠宇,王慧芳,吴向宏,等.基于依存句法分析的电力设备缺陷文本信息精确辨识方法[J].电力系统自动化,2020,44(12):178-185.
- [22]王海瑶,王慧芳,胡俊华,等.基于对话文本智能挖掘的电力设备供应商评价[J].电力自动化设备,2021,41(7):210-217.
- [23]邱剑,王慧芳,应高亮,等.文本信息挖掘技术及其在断路器全寿命状态评价中的应用[J].电力系统自动化,2016,40(6):107-112.
- [24] LIU Z Q,WANG H F. Automatic detection of transformer components in inspection images based on improved faster R-CNN[J].Energies,2018,11(12):3496-3513.
- [25]万吉林,吴国强,管敏渊,等.基于RetinaNet的变压器图像小部件智能识别方法[J].电力系统保护与控制,2021,49(12):166-173.
- [26]刘梓权,王慧芳,管敏渊,等.隔离开关图像数据扩充方法及其在自动状态识别中的应用[J].高电压技术,2020,46(2):441-447.
- [27]周志华.机器学习[M].北京:清华大学出版社,2016.
- 电力领域
- 数据驱动建模
- 新型电力系统
- 建模条件
- 建模风险
electric power domain - data-driven modeling
- new-type power system
- modeling conditions
- modeling risk