基于综合概率模型与深度学习的智能电网功率-电压映射方法

doi:10.12204/j.issn.1000-7229.2022.02.005

电力建设 ›› 2022, Vol. 43 ›› Issue (2): 37-44.doi: 10.12204/j.issn.1000-7229.2022.02.005

基于综合概率模型与深度学习的智能电网功率-电压映射方法

李建宜¹, 李鹏¹, 徐晓春²(), 施儒昱³, 曾平良⁴, 夏辉¹

1.华北电力大学电气与电子工程学院,河北省保定市 071003
2.国网江苏省电力有限公司淮安供电分公司,江苏省淮安市 223400
3.国网江苏省电力有限公司苏州供电分公司,江苏省苏州市 215000
4.杭州电子科技大学自动化学院,杭州市 310018

收稿日期:2021-09-23 出版日期:2022-02-01 发布日期:2022-03-24
通讯作者: 徐晓春 E-mail:877172717@qq.com
作者简介:李建宜(1996),男,硕士研究生,主要研究方向为人工智能在电力系统中的应用;
李鹏(1965),男,博士,博士生导师,教授,主要研究方向为新能源电力系统、综合能源系统与微网、能源互联网与智能电网、人工智能在智慧能源中的应用等;
施儒昱(1989),男,硕士,工程师,主要从事电力系统自动化方面的工作;
曾平良(1962),男,博士,教授,主要研究方向为电力系统分析与规化;
夏辉(1994),男,硕士研究生,主要研究方向为综合能源系统。
基金资助:
国家电网公司科技项目“含新能源、储能及柔性负荷的无功电压协调控制关键技术研究”(5108-202018028A-0-0-00)

Power-Voltage Mapping Method Based on Comprehensive Probability Model and Deep Learning for Smart Grid

LI Jianyi¹, LI Peng¹, XU Xiaochun²(), SHI Ruyu³, ZENG Pingliang⁴, XIA Hui¹

1. School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071003, Hebei Province, China
2. State Grid Huaian Power Supply Company, Huaian 223400, Jiangsu Province, China
3. State Grid Suzhou Power Supply Company, Suzhou 215000, Jiangsu Province, China
4. School of Automation, Hangzhou Dianzi University, Hangzhou 310018, China

Received:2021-09-23 Online:2022-02-01 Published:2022-03-24
Contact: XU Xiaochun E-mail:877172717@qq.com
Supported by:
Science and Technology Project of State Grid Corporation of China(5108-202018028A-0-0-00)

摘要/Abstract

摘要：

随着新能源及分布式发电渗透率的增加,其间歇性强、波动性大等特性对电网电压波动造成较大影响,如何更加快速地计算包含复杂分布式电源接入的电力系统稳态电压具有重要意义。通过对大量风机、光伏真实出力数据采样,在传统概率模型的基础上改进生成综合概率模型,并通过马尔科夫转移概率矩阵修正因时空特性产生的概率分布偏差。然后,以中国南方某地区相邻光伏、风电场实际出力数据为样本,基于配电网拓扑结构,在不同场景下计算其各节点稳态电压。最后,算例结果表明,改进方法模拟生成的电网模型具有较高真实性和适用性,计算所得的电压具有较高准确率。并且,相较传统电力系统潮流计算,极大减少计算时间,从而在控制效果上具有更好的跟随性,适用于复杂新能源电力系统稳态电压的计算。

关键词: 新能源, 综合概率模型, 马尔科夫链, 蒙特卡洛模拟, 人工智能

Abstract:

With the increase in the penetration rate of new energy based distributed power generation, its characteristics such as strong intermittent and high volatility will have a greater impact on grid voltage fluctuations. How to calculate the steady-state voltage of the power grid with complex distributed power access more quickly is of great significance. In this paper, by sampling a great deal of real output data of photovoltaic and wind power, a comprehensive probability model is improved and generated on the basis of the traditional probability model, and the probability distribution deviation caused by the temporal and spatial characteristics is corrected through the Markov transition probability matrix. Then, taking the actual output data of adjacent photovoltaic and wind power stations in a certain area of southern China as a sample, the steady-state voltage at each node of the distribution network under different scenarios is calculated in the distribution network topology. Finally, the results of the calculation example show that the power grid model generated by the improved method simulation in this paper has high authenticity and applicability, and the calculated voltages have high accuracy rate. Moreover, compared with the traditional power system flow calculation, the calculation time is greatly reduced, so that it has better follow-up in the control effect, and is suitable for the calculation of the steady-state voltage of the complex power system with new energy.

Key words: new energy, comprehensive probability model, Markov chain, Monte Carlo simulation, artificial intelligence

中图分类号:

TM743

李建宜, 李鹏, 徐晓春, 施儒昱, 曾平良, 夏辉. 基于综合概率模型与深度学习的智能电网功率-电压映射方法[J]. 电力建设, 2022, 43(2): 37-44.

LI Jianyi, LI Peng, XU Xiaochun, SHI Ruyu, ZENG Pingliang, XIA Hui. Power-Voltage Mapping Method Based on Comprehensive Probability Model and Deep Learning for Smart Grid[J]. ELECTRIC POWER CONSTRUCTION, 2022, 43(2): 37-44.

图/表 10

图1 功率-电压映射流程

Fig.1 Flowchart of power-voltage mapping

图2 光伏、风机综合概率模型建模方法

Fig.2 Flowchart of comprehensive probability density modeling of PV and wind

图3 相邻风机功率滞后示意

Fig.3 Power lag diagram of adjacent wind power generators

图4 基于深度学习的功率-电压映射方法

Fig.4 Power-voltage mapping method based on deep learning

图5 配电网拓扑结构

Fig.5 Topology of distribution network

表1 光伏接入点参数

Table 1 Parameters of photovoltaic access nodes

表2 风机接入点参数

Table 2 Parameters of wind power access nodes

图6 概率密度模型

Fig.6 Probability density model

图7 节点电压标准差

Fig.7 Standard deviation of node voltages

图8 神经网络训练结果

Fig.8 Neural network training

参考文献 23

[1]	LI P, GUO T, HAN X Q, et al. The optimal decentralized coordinated control method based on the H∞ performance index for an AC/DC hybrid microgrid[J]. International Journal of Electrical Power & Energy Systems, 2021, 125:106442. doi: 10.1016/j.ijepes.2020.106442 URL
[2]	LI P, GUO T, ZHOU F Q, et al. Nonlinear coordinated control of parallel bidirectional power converters in an AC/DC hybrid microgrid[J]. International Journal of Electrical Power & Energy Systems, 2020, 122:106208. doi: 10.1016/j.ijepes.2020.106208 URL
[3]	HAGHIFAM S, NAJAFI-GHALELOU A, ZARE K, et al. Stochastic bi-level coordination of active distribution network and renewable-based microgrid considering eco-friendly compressed air energy storage system and Intelligent parking lot[J]. Journal of Cleaner Production, 2021, 278:122808. doi: 10.1016/j.jclepro.2020.122808 URL
[4]	LI P, WANG Z X, WANG J H, et al. A multi-time-space scale optimal operation strategy for a distributed integrated energy system[J]. Applied Energy, 2021, 289:116698. doi: 10.1016/j.apenergy.2021.116698 URL
[5]	LI P, WANG Z X, LIU H T, et al. Bi-level optimal configuration strategy of community integrated energy system with coordinated planning and operation[J]. Energy, 2021, 236:121539. doi: 10.1016/j.energy.2021.121539 URL
[6]	刘健, 同向前, 潘忠美, 等. 考虑过电压因素时分布式光伏电源的准入容量[J]. 电力系统保护与控制, 2014, 42(6):45-51.
	LIU Jian, TONG Xiangqian, PAN Zhongmei, et al. The maximum power of distributed PV generation according to over-voltage in distribution network[J]. Power System Protection and Control, 2014, 42(6):45-51.
[7]	代景龙, 韦化, 鲍海波, 等. 基于无迹变换含分布式电源系统的随机潮流[J]. 电力自动化设备, 2016, 36(3):86-93.
	DAI Jinglong, WEI Hua, BAO Haibo, et al. Stochastic power flow calculation based on unscented transform for power system with distributed generations[J]. Electric Power Automation Equipment, 2016, 36(3):86-93.
[8]	王乃进, 韩松, 罗远国. 利用日最小负荷置信区间的光伏发电准入容量确定[J]. 电力系统及其自动化学报, 2020, 32(2):54-60.
	WANG Naijin, HAN Song, LUO Yuanguo. Method for determining the permitted capacity of photovoltaic generation using daily minimum load with confidence interval[J]. Proceedings of the CSU-EPSA, 2020, 32(2):54-60.
[9]	郑惠萍, 曾鹏, 刘新元, 等. 基于误差前馈预测的多时空尺度风电集群有功功率分层控制策略[J]. 电力建设, 2020, 41(8):120-128.
	ZHENG Huiping, ZENG Peng, LIU Xinyuan, et al. Active power layered control strategy based on error feedforward prediction for multiple temporal and spatial scales wind power cluster[J]. Electric Power Construction, 2020, 41(8):120-128.
[10]	黄伟, 葛良军, 华亮亮, 等. 基于概率潮流的主动配电网日前-实时两级优化调度[J]. 电力系统自动化, 2018, 42(12):51-57,105.
	HUANG Wei, GE Liangjun, HUA Liangliang, et al. Day-ahead and real-time optimal scheduling for active distribution network based on probabilistic power flow[J]. Automation of Electric Power Systems, 2018, 42(12):51-57,105.
[11]	汪惟源, 窦飞, 程锦闽, 等. 一种风光联合出力概率模型建模方法[J]. 电力系统保护与控制, 2020, 48(10):22-29.
	WANG Weiyuan, DOU Fei, CHENG Jinmin, et al. A modeling method for a wind and photovoltaic joint power probability model[J]. Power System Protection and Control, 2020, 48(10):22-29.
[12]	任洲洋, 颜伟, 项波, 等. 考虑光伏和负荷相关性的概率潮流计算[J]. 电工技术学报, 2015, 30(24):181-187.
	REN Zhouyang, YAN Wei, XIANG Bo, et al. Probabilistic power flow analysis incorporating the correlations between PV power outputs and loads[J]. Transactions of China Electrotechnical Society, 2015, 30(24):181-187.
[13]	杨挺, 韩旭涛, 姜含, 等. 电力系统科学计算在云数据中心的优化任务调配算法研究[J]. 电网技术, 2021, 12(27):1-13.
	YANG Ting, HAN Xutao, JIANG Han, et al. Research on task scheduling algorithm of power system scientific calculations in data center[J]. Power System Technology, 2021, 12(27):1-13.
[14]	杨虎, 刘琼荪, 钟波. 数理统计[M]. 北京: 高等教育出版社, 2004.
[15]	赵渊, 沈智健, 周念成, 等. 基于序贯仿真和非参数核密度估计的大电网可靠性评估[J]. 电力系统自动化, 2008, 32(6):14-19.
	ZHAO Yuan, SHEN Zhijian, ZHOU Niancheng, et al. Reliability assessment of bulk power systems utilizing sequential simulation and nonparametric kernel density estimation[J]. Automation of Electric Power Systems, 2008, 32(6):14-19.
[16]	任海军, 张晓星, 肖波, 等. 基于概念格的神经网络日最大负荷预测输入参数选择[J]. 吉林大学学报(理学版), 2011, 49(1):87-92.
	REN Haijun, ZHANG Xiaoxing, XIAO Bo, et al. Input parameters selection in neural network load forecasting mode based on concept lattice[J]. Journal of Jilin University (Science Edition), 2011, 49(1):87-92.
[17]	刘大贵, 王维庆, 张慧娥, 等. 马尔科夫修正的组合模型在新疆风电中长期可用电量预测中的应用[J]. 电网技术, 2020, 44(9):3290-3297.
	LIU Dagui, WANG Weiqing, ZHANG Huie, et al. Application of Markov modified combination model mid-long term available quantity of electricity forecasting in Xinjiang wind power[J]. Power System Technology, 2020, 44(9):3290-3297.
[18]	陈蓉珺, 何永秀, 陈奋开, 等. 基于系统动力学和蒙特卡洛模拟的电动汽车日负荷远期预测[J]. 中国电力, 2018, 51(9):126-134.
	CHEN Rongjun, HE Yongxiu, CHEN Fenkai, et al. Long-term daily load forecast of electric vehicle based on system dynamics and Monte Carlo simulation[J]. Electric Power, 2018, 51(9):126-134.
[19]	周艳真, 查显煜, 兰健, 等. 基于数据增强和深度残差网络的电力系统暂态稳定预测[J]. 中国电力, 2020, 53(1):22-31.
	ZHOU Yanzhen, ZHA Xianyu, LAN Jian, et al. Transient stability prediction of power systems based on deep residual network and data augmentation[J]. Electric Power, 2020, 53(1):22-31.
[20]	郑智聪, 王红, 齐林海. 基于深度学习模型融合的电压暂降源识别方法[J]. 中国电机工程学报, 2019, 39(1):97-104.
	ZHENG Zhicong, WANG Hong, QI Linhai. Recognition method of voltage sag sources based on deep learning models’ fusion[J]. Proceedings of the CSEE, 2019, 39(1):97-104.
[21]	余娟, 杨燕, 杨知方, 等. 基于深度学习的概率能量流快速计算方法[J]. 中国电机工程学报, 2019, 39(1):22-30.
	YU Juan, YANG Yan, YANG Zhifang, et al. Fast probabilistic energy flow analysis based on deep learning[J]. Proceedings of the CSEE, 2019, 39(1):22-30.
[22]	汤奕, 崔晗, 李峰, 等. 人工智能在电力系统暂态问题中的应用综述[J]. 中国电机工程学报, 2019, 39(1):2-13.
	TANG Yi, CUI Han, LI Feng, et al. Review on artificial intelligence in power system transient stability analysis[J]. Proceedings of the CSEE, 2019, 39(1):2-13.
[23]	闫龙川, 白东霞, 刘万涛, 等. 人工智能技术在云计算数据中心能量管理中的应用与展望[J]. 中国电机工程学报, 2019, 39(1):31-42.
	YAN Longchuan, BAI Dongxia, LIU Wantao, et al. Application and prospect of artificial intelligence technology in energy management and optimization for cloud computing data center[J]. Proceedings of the CSEE, 2019, 39(1):31-42.

基于综合概率模型与深度学习的智能电网功率-电压映射方法

Power-Voltage Mapping Method Based on Comprehensive Probability Model and Deep Learning for Smart Grid

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 23

相关文章 15

编辑推荐

Metrics

本文评价

[1]	刁利, 李光, 宋雪莹, 德格吉日夫, 谭忠富. 新能源分比例渗透下基于博弈的多主体效益综合评估方法[J]. 电力建设, 2022, 43(6): 43-55.
[2]	王萍萍, 许建中, 闫庆友, 林宏宇. 计及灵活性负荷资源需求响应和不确定性的楼宇微网调度双层优化模型[J]. 电力建设, 2022, 43(6): 128-140.
[3]	刘飞, 赵澄颢, 王世斌, 田旭, 刘联涛, 朱晓荣. 新能源直流汇集分群综合协调控制策略[J]. 电力建设, 2022, 43(5): 127-146.
[4]	杨立滨, 张磊, 刘艳章, 李正曦, 宗鸣. 基于分布式框架的新能源场站并网性能评估[J]. 电力建设, 2022, 43(5): 137-144.
[5]	康思伟, 董文凯, 郭诗然, 孙红军, 谢小荣. 基于虚拟同步机控制的新能源发电并网系统小干扰稳定临界短路比[J]. 电力建设, 2022, 43(3): 131-140.
[6]	刘一民, 王书扬, 李彬, 王兴国, 郑少明, 董鹏. 逆变型新能源场站柔性直流送出系统交流线路差动保护灵敏性优化方案[J]. 电力建设, 2022, 43(1): 63-69.
[7]	余洋, 贾浩, 陈启维, 范亚洲, 米增强, 袁玉宝, 常生强. 基于改进模型预测控制的电-气系统新能源功率波动平滑策略[J]. 电力建设, 2021, 42(9): 65-73.
[8]	李贤初, 张翕, 刘杰, 胡建林. 输电线路导线覆冰AMPSO-BP神经网络预测模型[J]. 电力建设, 2021, 42(9): 140-146.
[9]	张孟琛, 林丽娟, 孟菁, 牛益国, 王珺, 石磊磊. 数据驱动的配电台区源荷污染源群体谐波排放建模[J]. 电力建设, 2021, 42(8): 46-54.
[10]	何昭辉, 曹锐, 周成, 刘海鹏. 计及时空相关性的分布式风电接入配电网风险评估[J]. 电力建设, 2021, 42(7): 127-136.
[11]	任景, 薛晨, 马晓伟, 崔伟, 孟鑫羽, 杨迎, 王剑晓. 考虑能量存取的自备电厂去中心化交易机制探究[J]. 电力建设, 2021, 42(6): 1-8.
[12]	李晓萌, 张忠, 赵华, 饶宇飞, 张振安, 朱全胜, 吕泉. 能源互联网背景下电网备用问题探索[J]. 电力建设, 2021, 42(5): 57-68.
[13]	崔艳妍, 刘伟, 苏剑, 刘冀邱, 潘永超, 李嘉, 张先亮, 张璐. 考虑最优负荷削减与热负荷惯性的综合能源系统可靠性评估[J]. 电力建设, 2021, 42(4): 40-48.
[14]	张涛, 胡娱欧, 张晶, 韩亮, 王佩雯, 何淼, 陈雨果. 适应高比例新能源发展的电力市场容量保障机制分析[J]. 电力建设, 2021, 42(3): 117-125.
[15]	陈人杰, 李华取, 彭晓涛, 杨军, 董旭柱, 刘首文, 李祥杰. 基于改进TOPSIS的新能源大数据服务项目评价研究[J]. 电力建设, 2021, 42(3): 126-134.