Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 往年索引

基于参数自适应DBSCAN算法的旋转设备健康评估

来源:电工电气发布时间:2020-12-19 13:19 浏览次数:564
基于参数自适应DBSCAN算法的旋转设备健康评估
 
于凯,王哲,王玉龙,董恒章,刘宝楠,张世林
(安徽华电宿州发电有限公司,安徽 宿州 234000)
 
    摘 要:针对电厂旋转设备的运行状态异常检测问题,提出一种基于参数自适应DBSCAN算法的旋转设备健康状态在线评估算法。该算法中为降低人工设定邻域半径和密度阈值对密度聚类结果的影响,选用轮廓系数作为聚类结果有效性评价指标,基于粒子群算法(PSO)确定合理的参数值。采用参数自适应DBSCAN算法定期对正常运行时的历史数据进行离线聚类分析,基于此聚类结果分析实时采集的数据,在线评估旋转设备的健康指数。对某电厂旋转设备的运行数据进行仿真分析,结果表明所提方法能够有效检测设备异常运行状态,为设备的安全可靠运行提供保障。
    关键词:旋转设备;健康指数;参数自适应DBSCAN算法;粒子群算法;在线评估
    中图分类号:TM307     文献标识码:A     文章编号:1007-3175(2020)12-0024-06
 
Evaluation on Health of Rotation Equipment Based on Parameter Adaptive DBSCAN Algorithm
 
YU Kai, WANG Zhe, WANG Yu-long, DONG Heng-zhang, LIU Bao-nan, ZHANG Shi-lin
(Anhui Huadian Suzhou Power Generation Co., Ltd, Suzhou 234000, China)
 
    Abstract: In this paper, aiming at the detection of abnormal operation status of rotating equipment in power plants, this paper proposes an online health status assessment algorithm for rotating equipment based on parameter adaptive DBSCAN algorithm. In this algorithm, in order to reduce the influence of artificially set neighborhood radius (Eps) and density threshold (MinPts) on the results of density clustering, the contour coefficient is selected as Validity evaluation index of clustering results, determine reasonable parameter values based on particle swarm optimization (PSO). The parameter adaptive DBSCAN algorithm is used to periodically perform offline clustering analysis on historical data during normal operation. Based on this clustering result, the real-time collected data is analyzed, and the health index of the rotating equipment is evaluated online. After a simulation analysis of the operating data of a rotating equipment in a power plant, the results show that the proposed method can effectively detect the abnormal operating state of the equipment and provide a guarantee for the safe and reliable operation of the equipment.
    Key words: rotation equipment; health index; parameter adaptive DBSCAN algorithm; particle swarm optimization algorithm; online evaluation
 
参考文献
[1] 廖瑞金,王有元,刘航,等. 输变电设备状态评估方法的研究现状[J]. 高电压技术,2018,44(11):3454-3464.
[2] 江秀臣,盛戈皞. 电力设备状态大数据分析的研究和应用[J]. 高电压技术,2018,44(4):1041-1050.
[3] KANG Chongqing, WANG Yi, XUE Yusheng, et al.Big Data Analytics in China's Electric Power Industry: Modern Information, Communication Technologies, and Millions of Smart Meters[J]. IEEE Power and Energy Magazine,2018,16(3):54-65.
[4] 杨茂,杨琼琼. 基于云分段最优熵算法的风电机组异常数据识别研究[J]. 中国电机工程学报,2018,38(8):2294-2301.
[5] 田力, 向敏. 基于密度聚类技术的电力系统用电量异常分析算法[J]. 电力系统自动化,2017,41(5):64-70.
[6] 严英杰,盛戈皞,陈玉峰,等. 基于大数据分析的输变电设备状态数据异常检测方法[J]. 中国电机工程学报,2015,35(1):52-59.
[7] 程超,张汉敬,景志敏,等. 基于离群点算法和用电信息采集系统的反窃电研究[J]. 电力系统保护与控制,2015,43(17):69-74.
[8] 陈佳俊,陈玉峰,严英杰,等. 基于时空联合聚类方法的输变电设备状态异常检测[J]. 南方电网技术,2015,9(11):65-72.
[9] 邱志斌,阮江军,黄道春,等. 基于电机电流检测的高压隔离开关机械故障诊断[J]. 中国电机工程学报,2015,35(13):3459-3466.
[10] DREISBUSCH K, KRANZ H G, SCHNETTLER A. Determination of a failure probability prognosis based on PD-diagnostics in GIS[J]. IEEE Transactions on Dielectrics and Electrical Insulation,2008,15(6):1707-1714.
[11] 彭楚宁,罗冉冉,王晓东. 新一代智能电能表支撑泛在电力物联网技术研究[J]. 电测与仪表,2019,56(15):137-142.
[12] 窦健,刘宣,卢继哲,等. 基于用电信息采集大数据的防窃电方法研究[J]. 电测与仪表,2018,55(21):43-49.
[13] 李宁,尹小明,丁学峰,等. 一种融合聚类和异常点检测算法的窃电辨识方法[J]. 电测与仪表,2018,55(21):19-24.
[14] 李文杰,闫世强,蒋莹,等. 自适应确定DBSCAN算法参数的算法研究[J]. 计算机工程与应用,2019,55(5):1-7.
[15] 龙海侠,须文波,王小根,等. 基于选择操作的量子粒子群算法[J]. 控制与决策,2010,25(10):1499-1506.
[16] CLERC M, KENNEDY J.The Particle Swarm: Explosion, Stability, and Convergence in a Multi-Dimensional Complex Space[J].IEEE Transactions on Evolutionary Computation,2002,6(1):58-73.
[17] 王帅, 杜欣慧, 姚宏民, 等. 面向含多种用户类型的负荷曲线聚类研究[J]. 电网技术,2018,42(10):3401-3412.