基于APRIORI-贝叶斯优化XGBoost的电力通信网根告警预测

doi:10.12204/j.issn.1000-7229.2022.01.013

电力建设 ›› 2022, Vol. 43 ›› Issue (1): 113-121.doi: 10.12204/j.issn.1000-7229.2022.01.013

基于APRIORI-贝叶斯优化XGBoost的电力通信网根告警预测

程路明¹, 楼平¹, 诸骏豪¹, 李凌雁¹, 崔晓昱²(), 孙毅²

1.国网浙江省电力有限公司湖州供电公司,浙江省湖州市 313000
2.华北电力大学电气与电子工程学院,北京市 102206

收稿日期:2021-04-29 出版日期:2022-01-01 发布日期:2021-12-21
通讯作者: 崔晓昱 E-mail:524651088@qq.com
作者简介:程路明(1984),男,硕士研究生,高级工程师,主要研究方向为电力通信系统运行;
楼平(1967),男,硕士,高级工程师,主要研究方向为电力系统自动化;
诸骏豪(1990),男,学士,工程师,主要研究方向为电力通信;
李凌雁(1991),女,硕士,工程师,主要研究方向为电力通信;
孙毅(1972),男,博士,教授,主要研究方向为电力系统通信与信息技术。
基金资助:
国网浙江省电力有限公司科技项目“支撑电力物联网建设的信息通信网络综合性规划及其关键技术研究”(5211UZ190056)

Root Alarm Prediction of Power Communication Network Applying APRIORI-Bayesian Optimization XGBoost

CHENG Luming¹, LOU Ping¹, ZHU Junhao¹, LI Lingyan¹, CUI Xiaoyu²(), SUN Yi²

1. Huzhou Power Supply Company of State Grid Zhejiang Electric Power Co., Ltd., Huzhou 313000, Zhejiang Province, China
2. School of Electrical and Electronic Engineering, North China Electric Power University, Beijing 102206, China

Received:2021-04-29 Online:2022-01-01 Published:2021-12-21
Contact: CUI Xiaoyu E-mail:524651088@qq.com
Supported by:
Science and Technology Project of State Grid Zhejiang Electric Power Co., Ltd.(5211UZ190056)

摘要/Abstract

摘要：

电力通信网根源性告警的精准预测,能够辅助运维人员提前对通信网高风险点进行高效排查和快速定位,从根源上避免区域性通信故障和衍生告警,降低网络风险和运维成本。针对现有研究中电力通信网根告警预测源数据冗余、准确率不高的问题,面向电力通信网根告警提出基于APRIORI-贝叶斯优化XGBoost的预测模型,利用APRIORI算法优化预测模型输入,挖掘根告警影响因素间的关联规则,借助关联规则概率化方法确定关键影响因子,降低贝叶斯优化XGBoost模型训练数据冗余度,提高数据价值密度,进而提升模型效率和告警预测精度。实验结果表明,所提算法在预测准确率、召回率和F-值等性能上均取得良好的效果,并在最小支持度为15%时达到最优预测结果,能为电力通信网高效运维和故障排查提供技术支撑。

关键词: 根源性告警, 关联规则分析, 告警预测, 贝叶斯优化, XGBoost算法

Abstract:

The accurate prediction of the root alarms in the electric power communication network can assist the operation and maintenance personnel to efficiently investigate and quickly locate the high-risk points of the communication network in advance, then avoid regional communication failures and derivative alarms from the root, and reduce network risks and operation and maintenance costs. Aiming at the redundancy of source data and low accuracy of root alarm prediction in the existing research, this paper proposes a prediction model based on APRIORI-Bayesian optimization XGBoost for root alarms of electric power communication network. The APRIORI algorithm is used to optimize the input of the prediction model and mine the association rules among the influencing factors of root alarms. With the aid of the probabilistic method of association rules, the key influence factors are determined to reduce the training data redundancy of the Bayesian optimized XGBoost model, increase the data value density, and then improve the model efficiency and warning prediction accuracy. Then the prediction model is constructed on the basis of the Bayesian optimized XGBoost algorithm with the key factors. Finally, the experimental results show that the proposed algorithm performs well in prediction accuracy, recall and F-value, and achieves the optimal prediction accuracy when the minimum support is 15%, which can provide technical support for efficient maintenance and troubleshooting of power communication network.

Key words: root alarm, association rule analysis, alarm prediction, Bayesian optimization, XGBoost algorithm

中图分类号:

TM73
TN 915

程路明, 楼平, 诸骏豪, 李凌雁, 崔晓昱, 孙毅. 基于APRIORI-贝叶斯优化XGBoost的电力通信网根告警预测[J]. 电力建设, 2022, 43(1): 113-121.

CHENG Luming, LOU Ping, ZHU Junhao, LI Lingyan, CUI Xiaoyu, SUN Yi. Root Alarm Prediction of Power Communication Network Applying APRIORI-Bayesian Optimization XGBoost[J]. ELECTRIC POWER CONSTRUCTION, 2022, 43(1): 113-121.

图/表 14

表1 不同属性及其对应元素

Table 1 Different attributes and their corresponding elements

表2 随机抽取数据及元素

Table 2 Randomly extracted data and elements

表3 频繁1-项集L1

Table 3 Frequent 1-item set L1

表4 频繁2-项集L2

Table 4 Frequent 2-item set L2

表5 频繁3-项集L3

Table 5 Frequent 3-item set L3

表6 强关联规则挖掘结果分析

Table 6 Analysis of mining results of strong association rules%

图1 基于APRIORI-贝叶斯优化XGBoost的根告警预测流程图

Fig.1 Flowchart of root alarm prediction based on APRIORI-Bayesian optimization of XGBoost

表7 XGBoost参数范围及含义

Table 7 Range and meaning of XGBoost parameters

表8 根告警关联规则挖掘结果分析

Table 8 Analysis of the mining results of root alarm association rules

表9 不同算法的根告警预测性能分析

Table 9 Analysis of root alarm prediction performance of different algorithms

图2 关联因素挖掘对预测精度影响对比

Fig.2 Comparison of the impact of correlation factor mining on prediction accuracy

图3 4种算法预测结果对比

Fig.3 Comparison of prediction results of 4 algorithms

图4 不同告警数据量下算法执行时间

Fig.4 Algorithm execution time under different alarm data volume

图5 不同支持度下告警预测性能

Fig.5 Alarm prediction performance under different support levels

参考文献 22

[1]	刘林, 祁兵, 李彬, 等. 面向电力物联网新业务的电力通信网需求及发展趋势[J]. 电网技术, 2020, 44(8): 3114-3130.
	LIU Lin, QI Bing, LI Bin, et al. Requirements and developing trends of electric power communication network for new services in electric Internet of Things[J]. Power System Technology, 2020, 44(8): 3114-3130.
[2]	汤亿则, 庄毅. 电力通信网根告警诊断的原理与研究[J]. 电力系统通信, 2011, 32(11): 9-12.
	TANG Yize, ZHUANG Yi. Principles and research on root alarm diagnosing in power communication network[J]. Telecommunications for Electric Power System, 2011, 32(11): 9-12.
[3]	杨志敏, 洪丹轲, 黄昱. 电力通信网省地集约化监控运行支撑系统架构及关键技术[J]. 电信科学, 2019, 35(11): 146-156.
	YANG Zhimin, HONG Danke, HUANG Yu. Architecture and key technologies of the operation support system for intensive monitoring of provincial and local power communication network[J]. Telecommunications Science, 2019, 35(11): 146-156.
[4]	YAN B, ZHAO Y, RAHMAN S, et al. Dirty-data-based alarm prediction in self-optimizing large-scale optical networks[J]. Optics Express, 2019, 27(8): 10631-10643. doi: 10.1364/OE.27.010631 URL
[5]	刘军, 王颖, 张冰, 等. 电力骨干通信网中告警预测方法研究[J]. 光通信研究, 2019(5): 9-13.
	LIU Jun, WANG Ying, ZHANG Bing, et al. Research on alarm prediction method based on data pre-processing in power backbone communication network[J]. Study on Optical Communications, 2019(5): 9-13.
[6]	齐小刚, 胡秋秋, 姚旭清, 等. 一种有效的通信网络告警分析方法[J]. 西安电子科技大学学报, 2019, 46(4): 1-8.
	QI Xiaogang, HU Qiuqiu, YAO Xuqing, et al. Efficient alarm analysis approach for communication networks[J]. Journal of Xidian University, 2019, 46(4): 1-8.
[7]	王帅, 杨秋辉, 曾嘉彦, 等. 加权增量关联规则挖掘在通信告警预测中的应用[J]. 计算机应用, 2018, 38(10): 2875-2880. doi: 10.11772/j.issn.1001-9081.2018020392
	WANG Shuai, YANG Qiuhui, ZENG Jiayan, et al. Application of weighted incremental association rule mining in communication alarm prediction[J]. Journal of Computer Applications, 2018, 38(10): 2875-2880. doi: 10.11772/j.issn.1001-9081.2018020392
[8]	姜文, 王禄海, 吴清玉, 等. 电力系统通信网络状态估计模型与智能告警[J]. 电力系统及其自动化学报, 2019, 31(2): 101-105.
	JIANG Wen, WANG Luhai, WU Qingyu, et al. State estimation model for power system communication network and intelligent alarm[J]. Proceedings of the CSU-EPSA, 2019, 31(2): 101-105.
[9]	LIN P, YE K J, CHEN M, et al. DCSA: using density-based clustering and sequential association analysis to predict alarms in telecommunication networks[C]// 2019 IEEE 25th International Conference on Parallel and Distributed Systems (ICPADS). December 4-6, 2019, Tianjin, China. IEEE, 2019: 1-8.
[10]	杨剑, 蓝明超. 基于业务时间窗选取的告警聚类及关联方法[J]. 光通信研究, 2019(1): 33-36.
	YANG Jian, LAN Mingchao. A method for alarm clustering and correlation based on business time window selection[J]. Study on Optical Communications, 2019(1): 33-36.
[11]	ZHANG M, WANG D S. Machine learning based alarm analysis and failure forecast in optical networks[C]// 2019 24th OptoElectronics and Communications Conference (OECC) and 2019 International Conference on Photonics in Switching and Computing (PSC). July 7-11, 2019, Fukuoka, Japan. IEEE, 2019: 1-3.
[12]	ZHUANG H T, ZHAO Y L, YU X S, et al. Machine-learning-based alarm prediction with GANs-based self-optimizing data augmentation in large-scale optical transport networks[C]// 2020 International Conference on Computing, Networking and Communications (ICNC). February 17-20, 2020, Big Island, HI, USA. IEEE, 2020: 294-298.
[13]	LI T Y, CHEN C. Alarm sequences forecasting based on sparse Bayesian in communication networks[C]// 2011 International Conference on Mechatronic Science, Electric Engineering and Computer (MEC). August 19-22, 2011, Jilin, China. IEEE, 2011: 2233-2236.
[14]	ZHANG B, ZHAO Y L, RAHMAN S, et al. Alarm classification prediction based on cross-layer artificial intelligence interaction in self-optimized optical networks (SOON)[J]. Optical Fiber Technology, 2020, 57: 102251.
[15]	陈勇, 李胜男, 张丽, 等. 基于改进Apriori算法的智能变电站二次设备缺陷关联性分析[J]. 电力系统保护与控制, 2019, 47(20): 135-141.
	CHEN Yong, LI Shengnan, ZHANG Li, et al. Association analysis for defect data of secondary device in smart substations based on improved Apriori algorithm[J]. Power System Protection and Control, 2019, 47(20): 135-141.
[16]	WANG H, SUN F J. Optimal sensor placement and fault diagnosis model of PV array of photovoltaic power stations based on XGBoost[J]. IOP Conference Series: Earth and Environmental Science, 2021, 661(1): 012025.
[17]	徐锋, 方彦军. 基于贝叶斯优化XGBoost的现场校验仪误差预测[J]. 电测与仪表, 2019, 56(18): 120-125.
	XU Feng, FANG Yanjun. Error prediction of field calibrator based on Bayesian optimization XGBoost[J]. Electrical Measurement & Instrumentation, 2019, 56(18): 120-125.
[18]	魏乐, 胡晓东, 尹诗. 基于优化XGBoost的风电机组发电机前轴承故障预警[J]. 系统仿真学报, 2021, 33(10): 2335-2343. doi: 10.16182/j.issn1004731x.joss.20-0538
	WEI Le, HU Xiaodong, YIN Shi. Optimized-XGBoost early warning of wind turbine generator front bearing fault[J]. Journal of System Simulation, 2021, 33(10): 2335-2343.
[19]	崔佳旭, 杨博. 贝叶斯优化方法和应用综述[J]. 软件学报, 2018, 29(10): 3068-3090.
	CUI Jiaxu, YANG Bo. Survey on Bayesian optimization methodology and applications[J]. Journal of Software, 2018, 29(10): 3068-3090.
[20]	张光兰, 杨秋辉, 程雪梅, 等. 序列模式挖掘在通信网络告警预测中的应用[J]. 计算机科学, 2018, 45(S2): 535-538.
	ZHANG Guanglan, YANG Qiuhui, CHENG Xuemei, et al. Application of sequence pattern mining in communication network alarm prediction[J]. Computer Science, 2018, 45(S2): 535-538.
[21]	谭武坤, 杨秋辉, 陈伟. 基于贝叶斯网络的通信网络故障定位方法[J]. 计算机应用, 2018, 38(S2): 217-220.
	TAN Wukun, YANG Qiuhui, CHEN Wei. Fault location method of communication network based on Bayesian network[J]. Journal of Computer Applications, 2018, 38(S2): 217-220.
[22]	邱少明, 杨雯升, 杜秀丽, 等. 优化随机森林模型的网络故障预测[J]. 计算机应用与软件, 2021, 38(2): 103-109.
	QIU Shaoming, YANG Wensheng, DU Xiuli, et al. Network fault prediction based on improved random forest model[J]. Computer Applications and Software, 2021, 38(2): 103-109.

基于APRIORI-贝叶斯优化XGBoost的电力通信网根告警预测

Root Alarm Prediction of Power Communication Network Applying APRIORI-Bayesian Optimization XGBoost

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 22

相关文章 15

编辑推荐

Metrics

本文评价

[1]	闫庆友, 史超凡, 秦光宇, 许传博. 基于近端策略优化算法的电化学/氢混合储能系统双层配置及运行优化[J]. 电力建设, 2022, 43(8): 22-32.
[2]	杨秀, 焦楷丹, 孙改平, 陈小毅, 杜佳玮, 仇志鑫. 考虑负荷多无功用电场景的城市配电网无功优化配置[J]. 电力建设, 2022, 43(8): 42-52.
[3]	李振坤, 王艺龙, 季亮, 张智泉. 基于虚拟变电站的配电网薄弱区域集群双层规划研究[J]. 电力建设, 2022, 43(8): 53-65.
[4]	詹祥澎, 杨军, 沈一民, 钱晓瑞, 王昕妍, 吴赋章. 基于价值认同的需求侧电能共享分布式交易策略[J]. 电力建设, 2022, 43(8): 141-149.
[5]	李军徽, 马得轩, 朱星旭, 李翠萍, 侯涛. 基于ADMM算法的主动配电网分层优化经济调度[J]. 电力建设, 2022, 43(8): 76-86.
[6]	吴界辰, 汪莹, 易海琼, 李晖, 刘建琴, 冯腾. 中国台湾“3·3”大规模停电事故分析及其对未来电网发展的启示[J]. 电力建设, 2022, 43(7): 73-79.
[7]	董朝武, 李哲, 蒲莹, 钟启迪, 李明, 郝俊芳. 一种应用于柔性直流输电控制保护系统的波特率动态自适应通信技术[J]. 电力建设, 2022, 43(7): 57-62.
[8]	宫建锋, 周宗川, 靳盘龙, 张斌, 韩一鸣, 冯雪. 考虑灵活性的交直流混合配电网储能双层规划[J]. 电力建设, 2022, 43(7): 48-56.
[9]	李鹏, 余晓鹏, 张艺涵, 周青青, 田春筝, 乔慧婷. 计及碳捕集和电转气的农村虚拟电厂多目标随机调度优化模型[J]. 电力建设, 2022, 43(7): 24-36.
[10]	时维帅, 孙欣, 谢敬东, 严佳嘉, 张誉匀. “双碳”目标下风氢联合系统参与现货市场的优化运行与效益分析[J]. 电力建设, 2022, 43(7): 1-12.
[11]	陈泽西, 王朴, 肖万芳, 贾东强, 耿欣扬, 杨立. 计及可靠性收益的配电网柔性多状态开关及分布式电源综合优化配置[J]. 电力建设, 2022, 43(6): 93-100.
[12]	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.
[13]	李晓刚, 杨立兵, 吴敏, 万家豪, 冯冬涵, 周云. 省级现货市场出清结果对区域同步电网潮流越限的影响[J]. 电力建设, 2022, 43(6): 34-42.
[14]	杨雪, 金孝俊, 王海洋, 魏翼飞. 基于区块链的绿证和碳交易市场联合激励机制[J]. 电力建设, 2022, 43(6): 24-33.
[15]	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.