随着风电并网容量的不断攀升，电力系统的不确定性不断增强，给电力系统稳定运行带来了新的挑战。同时近年来多起大停电事故结果表明，高比例新能源并网后，电网的自调节能力和抗扰动能力明显改变，系统更易发生连锁故障。研究发现，事故大多是由于一个或多个正在运行的元件发生故障导致其他线路过载跳闸、电压崩溃，引起大范围的潮流转移，继而引发连锁故障。因此，揭示连锁故障发生机理，建立系统关键线路识别指标及连锁故障仿真模型，对风电并网情况下的关键线路识别及连锁故障预测等问题进行研究，对于预防大停电事故、维持电力系统稳定运行具有重大意义。论文的主要内容如下：

首先，基于Copula函数建立多风电场功率出力的相关性模型，并从三个不同的角度来选取最优Copula函数模型。以Elia电网的陆上风电场Elia-Connected Wind Farms和DSO-Connected Wind Farms的2022年6至8月风机实际出力数据为例进行分析，结果表明使用t-Copula函数能够描述两风电场的出力相关特性。通过拉丁超立方抽样对Copula函数模型的多风电场功率数据进行数据削减，为进行含风电场的关键线路识别及连锁故障预测奠定基础。

其次，为解决高比例风电并网后，现有的确定性关键线路辨识方法无法识别考虑风电并网功率不确定情况下的关键线路，本文提出了一种新的风电并网情况下的关键线路辨识方法。在风电并网情况下，建立了识别关键线路的电压稳定性、线路负载率和拓扑关联度3个指标，通过最优潮流以及组合赋权法的计算结果，计算每条线路的综合重要性指标来辨识关键线路。以IEEE39节点系统为例，考虑风电并网情况下风机出力的不确定性和相关性，对所提出的关键线路识别方法进行测试。结果表明，所提方法可以识别不同风电并网条件下的关键线路，可为预防大停电事故提供参考。

最后，建立风电并网情况下的连锁故障的预测模型。通过最优潮流计算风电并网情况下初始故障指标值，将指标值较高的线路列为初始故障集，然后断开所选线路进行后续连锁故障的搜索并判定预测是否结束，最后进行连锁故障事故集的生成。在上述建立的连锁故障预测模型基础上，运用Eclat算法对连锁故障事故链中的数据进行挖掘。在案例分析中对比风电并网两种情况下IEEE-39节点系统的频繁线路集，结果表明所提出的连锁故障预测模型能够有效地预测风电并网情况下的连锁故障。

With the continuous increase of wind power grid-connected capacity, the uncertainty of the power system continues to increase, which brings new challenges to the stable operation of the power system. At the same time, the results of many large blackouts in recent years show that after a high proportion of new energy is connected to the grid, the self-regulating ability and anti-disturbance ability of the grid will change significantly, and the system will be more prone to cascading failures. Studies have found that most of the accidents are caused by the failure of one or more operating components, causing overload tripping and voltage collapse of other lines, causing large-scale power flow transfer, and then causing cascading failures. Therefore, the study of the mechanism of cascading failures, the establishment of system key line identification indicators and cascading failure simulation models, and the research on key line identification and cascading failure predictions in the case of wind power grid-connected are important for preventing blackouts and maintaining stable operation of the power system. has great significance. The main content of the paper is as follows:

First, the correlation model of multi-wind farm power output is established based on the Copula function, and the optimal Copula function model is selected from three different angles. Taking the actual wind turbine output data of Elia-Connected Wind Farms and DSO-Connected Wind Farms of Elia Grid from June to August 2022 as an example, the results show that the output correlation characteristics of the two wind farms can be described by using the t-Copula function. The multi-wind farm power data of the Copula function model is reduced by Latin hypercube sampling, which lays the foundation for the identification of key lines including wind farms and the prediction of cascading failures.

Secondly, in order to solve the problem that the existing deterministic key line identification method cannot identify the key line in the case of uncertain wind power grid-connected power after a high proportion of wind power is connected to the grid, this paper proposes a new key line in the case of wind power grid-connected identification method. In the case of wind power grid-connected, three indicators are established to identify key lines, voltage stability, line load rate, and topology correlation. Using the calculation results of optimal power flow and combined weighting method, the comprehensive importance index of each line is calculated. to identify critical lines. Taking the IEEE39 node system as an example, considering the uncertainty and correlation of wind power output in the case of wind power grid-connected, the proposed key line identification method is tested. The results show that the key lines under different wind power grid-connected conditions can be identified. Provide reference for preventing blackout accidents.

Finally, a prediction model for cascading failures in the case of wind power grid-connected is established: calculate the initial fault index value in the case of wind power grid-connected through the optimal power flow, list the lines with higher index values as the initial fault set, and then disconnect the selected lines for further analysis. Search for subsequent cascading failures and determine whether the prediction is over, and finally generate cascading failure accident sets. On the basis of the cascading failure prediction model established above, the data in the cascading failure accident chain is mined by using the Eclat algorithm. In the case analysis, comparing the frequent line sets of the IEEE-39 node system in two cases of wind power grid-connected, the results show that the proposed cascading failure prediction model can effectively predict the cascading failures in the case of wind power grid-connected.

[1]国家发展改革委, 国家能源局. 2022年全国电力工业统计数据[R]. 北京: 国家发展改革委, 国家能源局, 2023．

[2]习近平. 继往开来, 开启全球应对气候变化新征程[N]. 人民日报, 2020-12-13(002).

[3]申洪, 周勤勇, 刘耀, 等. 碳中和背景下全球能源互联网构建的关键技术及展望[J]. 发电技术, 2021, 42(01): 8-19.

[4]周孝信. 能源转型中我国新一代电力系统技术发展趋势[J]. 电气时代,2018, (01): 33-35.

[5]董新洲, 汤涌, 卜广全, 等. 大型交直流混联电网安全运行面临的问题与挑战[J]. 中国电机工程学报, 2019, 39(11): 3107- 3119.

[6]林伟芳, 易俊, 郭强, 等. 阿根廷“6.16”大停电事故分析及对中国电网的启示[J]. 中国电机工程学报, 2020, 40(09): 2835- 2842.

[7]孙华东, 许涛, 郭强, 等. 英国“8·9”大停电事故分析及对中国电网的启示[J]. 中国电机工程学报, 2019, 39(21): 6183-6192.

[8]曾辉, 孙峰, 李铁, 等. 澳大利亚“9·28”大停电事故分析及对中国启示[J]. 电力系统自动化, 2017, 41(13): 1-6.

[9]李志强. 电力调度自动化系统定向攻击的虚构陷阱主动安全防护[D]. 长沙: 长沙理工大学,2017.

[10]孙晓伟. 基于有功潮流介数的电网脆弱环节辨识研究[D]. 宜昌: 三峡大学, 2016.

[11]张涛, 孙晓伟, 徐雪琴, 等. 基于有功潮流介数的电网关键线路辨识[J]. 电网技术, 2016, 40(01): 193-198.

[12]田宇, 彭志炜, 王杰, 等. 基于特征值灵敏度的D-PMSG并网系统次同步振荡特性研究[J]. 节能技术, 2020, 38(05): 399-406.

[13]张赵青,朱建国,王雪,等.影响风电场电能质量的因素研究[J].节能技术,2014,32(06):516-519.

[14]郭涛, 秦睿, 马呈霞, 等. 利用回热抽汽控制参与风电系统调频方法研究[J]. 节能技术, 2014, 32(02): 108-111.

[15]张鹏飞, 麻常辉, 李威, 等. 2021年欧洲电网两次解列事故分析及对中国电网安全的思考[J]. 电力系统自动化, 2021, 45 (24):22-29.

[16]丁岩, 郭和强, 王利利, 等. 基于马尔可夫模型的电力系统连锁故障预测[J]. 电网与清洁能源, 2017, 33(09): 93-98.

[17]马志远, 石立宝, 姚良忠, 等. 电网连锁故障的事故链搜索模型及策略研究[J]. 中国电机工程学报, 2015, 35(13): 3292- 3302.

[18]赵光远, 吴旭东, 刘友波. 基于改进GN算法的电网关键输电断面辨识方法[J]. 智慧电力, 2021, 49(09): 70-76.

[19]林攀, 吴佳毅, 黄涛, 等. 电力系统脆弱性评估综述[J]. 智慧电力, 2021, 49(01): 22-28.

[20]杨茂, 杜刚. 基于t Location-Scale分布的风电功率概率预测研究[J]. 中国电力, 2017, 50(01): 140-145.

[21]Idriss A.I.,Mohamed A.A., Akinci T.C., et al. Accuracy of Eight Probability Distribution Functions for Modeling Wind Speed Data in Djibouti[J]. International Journal of Renewable Energy Research, 2020, 10(2): 780-790.

[22]Ouahabi M. H., Benabdelouahab F., Khamlichi A. Analyzing wind speed data and wind power density of Tetouan city in Morocco by adjustment to Weibull and Rayleigh distribution functions[J]. Wind Engineering, 2017, 41(3): 174-184.

[23]樊扬, 林勇, 徐乾耀, 等. 广东电网风电出力特性分析及其经济性评价[J]. 南方电网技术, 2012, 6(01): 8-12.

[24]叶林, 张亚丽, 巨云涛, 等. 用于含风电场的电力系统概率潮流计算的高斯混合模型[J]. 中国电机工程学报, 2017, 37(15): 4379-4387+4578.

[25]杨茂, 董骏城. 基于混合分布模型的风电功率波动特性研究[J]. 中国电机工程学报, 2016, 36(S1): 69-78.

[26]张帅可, 罗萍萍. 基于混合分布模型的风电功率超短期预测误差分析[J]. 电力科学与技术学报, 2020,35(05): 111-118.

[27]黄煜, 徐青山, 卞海红, 等. 基于拉丁超立方采样技术的半不变量法随机潮流计算[J]. 电力自动化设备, 2016, 36(11): 112-119.

[28]Chen Y, Wen J, Cheng S. Probabilistic Load Flow Method Based on Nataf Transformation and Latin Hypercube Sampling[J].IEEE Transactions on Sustainable Energy, 2013,4(2): 294-301．

[29]赵继超, 袁越, 傅质馨, 等. 基于Copula理论的风光互补发电系统可靠性评估[J]. 电力自动化设备, 2013, 33(01): 124-129.

[30]谢敏, 熊靖, 刘明波, 等. 基于Copula的多风电场出力相关性建模及其在电网经济调度中的应用[J]. 电网技术, 2016, 40(04): 1100-1106.

[31]Xie Z Q, Ji T Y, Li M S, et al. Quasi-Monte Carlo Based Probabilistic Optimal Power Flow Considering the Correlation of Wind Speeds Using Copula Function[J].IEEE Transactions on Power Systems,2017,33(2):2239-2247.

[32]Sun C, Bie Z, Xie M, et al. Fuzzy copula model for wind speed correlation and its application in wind curtailment evaluation[J]．Renewable Energy,2016,93(1):68-76.

[33]曹一家, 王光增, 曹丽华， 等. 基于潮流熵的复杂电网自组织临界态判断模型[J]. 电力系统自动化, 2011, 35(07): 1-6.

[34]张弘历, 李华强, 杨植雅, 等. 基于潮流增长率泰尔熵的脆弱支路辨识[J]. 电网技术, 2017, 41(07): 2340-2346.

[35]刘威, 张东霞, 丁玉成, 等.基于随机矩阵理论与熵理论的电网薄弱环节辨识方法[J]. 中国电机工程学报, 2017, 37(20): 5893-5901.

[36]吴旭. 基于N-k故障的电力系统运行风险及脆弱性评估[D]. 北京: 华北电力大学, 2013.

[37]王龙. 电力系统安全风险与脆弱性分析评估方法[D]. 北京: 华北电力大学, 2017.

[38]骆婷婷. 电网关键线路辨识及连锁故障风险评估研究[D]. 合肥: 合肥工业大学, 2018.

[39]邢琦琳, 张元标, 梁志宁, 等. Dynamics of organizational rumor communication on connecting multi-small-world networks[J]. Chinese Physics B, 2011, 20(12): 35-41.

[40]高伟. 基于无标度和小世界特性的网络重要节点识别[D]. 西安: 西安电子科技大学, 2019.

[41]刘耀年, 术茜, 康科飞, 等. 基于电抗加权介数指标的电网脆弱线路识别[J]. 电力系统保护与控制, 2011, 39 (23): 89-92 +100.

[42]徐林, 王秀丽, 王锡凡. 电气介数及其在电力系统关键线路识别中的应用[J]. 中国电机工程学报, 2010, 30(01): 33-39.

[43]刘文颖, 梁才, 徐鹏, 等. 基于潮流介数的电力系统关键线路辨识[J]. 中国电机工程学报, 2013, 33(31): 90-98+11

[44]陶鸿飞, 谢栋, 赵福林, 等. 含大规模风电并网电力系统的关键线路综合辨识[J]. 电力系统保护与控制, 2020, 48(06): 115-123.

[45]徐敬友, 陈冲, 罗纯坚, 等. 基于改进复杂网络模型的电网关键环节辨识[J]. 电力系统自动化, 2016, 40(10): 53-61.

[46]宗祥瑞. 基于熵权法的电网运行状态特征分析研究[D]. 天津: 天津大学,2017.

[47]张宾. 基于改进Matpower连续潮流算法的静态电压稳定性分析[J]. 西安文理学院学报(自然科学版), 2020, 23(02): 25-29.

[48]李亚婷. 含风电场的电力系统静态电压稳定性分析[D]. 秦皇岛: 燕山大学,2017.

[49]North American Electric Reliability Council. NERC planning standards[Z]. Glossary of Terms Used in Reliability Standards, 2008.

[50]R Baldick, B Chowdhury, I Dobson, et al. Initial review of methods for cascading failure analysis in electric power transmission systems [C]// Power and Energy Society General Meeting - Conversion and Delivery of Electrical Energy in the Century, 2008:1-8.

[51]钱宇骋, 陈忠, 杨为, 等. 电力系统连锁故障研究综述[J]. 山东电力技术, 2018, 45(12): 1-9.

[52]Saeed H, Hossein F-A, Esmaeil M-M. Power system security improvement using an OPA model and IPSO algorithm[J]. SIMULATION,2020,96(3).

[53]何晓峰, 黎寿涛, 夏成军, 等. 基于OPA连锁故障模型的电力系统风险评估[J]. 南方电网技术, 2019, 13(10): 31-38.

[54]Wu X.,Wu D.,Modiano E.. An Influence Model Approach to Failure Cascade Prediction in Large Scale Power Systems[J]. Proceedings of the American Control Conference , 2020,2020-July.

[55]周健. 计及风电接入的电网时空演化模型及连锁故障研究[D]. 湘潭: 湘潭大学, 2019.

[56]I Dobson, K R Wierzbicki, J Kim, et al. Towards quantifying cascading blackout risk[C]// 2007 iREP Symposium-Bulk Power System Dynamics and Control, USA, 2007.

[57]赵广旭, 穆小亮, 黄小燕. 基于继电保护隐性故障的电力系统连锁故障分析[J]. 信息系统工程, 2019(04): 109.

[58]赵炜炜, 张建华, 易俊, 等.改进的Manchester连锁故障模型及其应用[J]. 电网技术, 2009, 33(11): 18-22.

[59]申原. 一种计及继电保护系统多模式隐藏故障的电网N-k风险方法研究[J]. 自动化与仪器仪表, 2015(11): 171-172.

[60]L Z Li, L Liu, H Wu, et al. Identification of Critical Hidden Failure Line Based on State-failure-network[J].Journal of Modern Power Systems and Clean Energy, 2022, 10(01): 40-49.

[61]朱天曈, 丁坚勇, 田世明, 等. 基于复合脆性关联度与改进Floyd算法的连锁故障预测[J]. 电力科学与技术学报, 2018, 33(04): 58-65+72.

[62]李庆. 电网关键线路及连锁故障辨识方法研究[D]. 吉林: 东北电力大学, 2016.

[63]周强明, 姜盛波, 鲁鸿毅, 等. 基于FP-growth算法的大电网关键线路辨识方法[J]. 电力自动化设备, 2018, 38(04): 89-95.

[64]吴旭, 张建华, 吴林伟, 等. 考虑风电场接入的电网连锁故障运行风险评估[J]. 华东电力, 2013, 41(02): 394-400.

[65]邱紫阳. 考虑风电不确定性的电网连锁故障建模研究[D]. 成都: 西南交通大学,2020.

[66]张雪敏, 谢宇翔, 郭金鹏, 等. 考虑等级事故概率约束的输电网利用效率评估方法[J]. 电力系统自动化, 2017, 41(07): 1-7.