考虑到边坡安全监测数据中存在粗差这一问题,提出了一种基于奇异谱分析(SSA)和密度聚类算法(DBSCAN)的粗差探测法,该方法结合SSA在提取信号和DBSCAN算法在区分粗差和异常值上的优势:首先使用SSA对监测序列进行分解重构,准确提取主信号并获取残余分量;然后使用DBSCAN聚类算法对残余分量进行分析;最后联合2种方法确定粗差点并剔除。通过引入多因素影响的边坡监测序列实例进行验证,并且将SSA-DBSCAN粗差探测法与中位数绝对偏差法(MAD)和格拉布斯准则法(Grubbs)进行比较分析。结果表明,本文提出的SSA-DBSCAN粗差探测法与上述方法相比性能优异、误判率低,可为后续监测数据分析处理乃至于预测预警奠定基础。
Abstract
A method of detecting the gross error of slope monitoring data is presented based on singular spectrum analysis (SSA) and density-based spatial clustering of applications with noise (DBSCAN). The method integrates the advantages of SSA in signal extraction and DBSCAN in distinguishing gross errors and outliers. Firstly, SSA is used to decompose and reconstruct the monitoring series to accurately extract the main signal and obtain the residual components. Secondly, DBSCAN is employed to analyze the residual components. The two methods are combined to determine and eliminate the gross errors. Examples of slope monitoring series affected by multiple factors are introduced for verification. Moreover, the present method is compared with the median absolute deviation method (MAD) and Grubbs criterion method (Grubbs), and results suggest that the present SSA-DBSCAN method is of excellent performance and low misjudgment rate compared with the abovementioned methods.
关键词
边坡工程 /
奇异谱分析 /
时间序列 /
安全监测数据 /
粗差探测 /
DBSCAN
Key words
slope engineering /
singular spectrum analysis /
time series /
safety monitoring data /
gross error detection /
DBSCAN
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 孙辅庭.大坝安全监测数据粗差智能识别技术研究[C] //中国水力发电工程学会大坝安全专委会2019年会暨“大坝安全现场检查评估与补强加固”学术交流会论文集.普洱:中国水力发电工程学会大坝安全专委会,2019:79-84.
[2] 吕玉红.时间序列异常检测算法的研究与应用[D].成都:电子科技大学,2018.
[3] 杨 哲,李艳玲,张 鹏,等.基于M估计量及标准四分位间距的安全监测数据异常识别的改进方法[J].长江科学院院报,2020,37(6):77-80.
[4] 蔡晓军,杨建华.基于多通道奇异谱的GNSS坐标序列粗差探测与数据插值[J].测绘工程,2019,28(5):20-28,34.
[5] 张东华,李志娟,刘全明,等.奇异谱分析的变形监测序列粗差探测方法[J].测绘科学,2020,45(8):14-18.
[6] 陈利军,王 畅.基于DBSCAN的地震电离层扰动异常数据检测方法[J].地震工程学报,2020,42(2):410-415.
[7] 罗怡澜,邹益胜,王 超,等.基于K-means-DBSCAN融合聚类的轴承异常温升诊断模型[J].机械设计与制造,2020(3):18-23.
[8] 郑霞忠,陈国梁,邹 韬.考虑时间序列关联的大坝监测异常数据清洗[J].水力发电,2020,46(4):111-114,125.
[9] 王 露. 大坝施工期温度监测数据关联规则挖掘与预测[D].北京:清华大学,2017.
[10] 王解先,连丽珍,沈云中.奇异谱分析在GPS站坐标监测序列分析中的应用[J].同济大学学报(自然科学版),2013,41(2):282-288.
[11] 李世友,王奉伟,沈云中.大坝变形时间序列的奇异谱分析[J].测绘通报,2018(9):64-68.
[12] COLEBROOK J M.Continuous Plankton Records: Zooplankton and Environment,Northeast Atlantic and North-Sea, 1948-1975[J]. Oceanologica Acta,1978,1(1): 9-23.
[13] 梁小珍,郭战坤,张倩文,等.基于奇异谱分析的航空客运需求分析与分解集成预测模型[J].系统工程理论与实践,2020,40(7):1844-1855.
[14] 卢辰龙.奇异谱分析在大地测量时间序列分析中的应用研究[D].长沙:中南大学,2014.
[15] 孔 雯,车 权,赵慧荣,等.基于奇异谱分析与长短时记忆神经网络的电厂存煤量短期预测[J].信息与控制,2020,49(6):742-751.
[16] 毛鹏宇,陈 义,孟 鑫.基于奇异谱分析的地基GPS反演可降水量可行性分析[J].大地测量与地球动力学,2017,37(9):933-936.
[17] SANEI S, LEE T K M, ABOLGHASEMI V. A New Adaptive Line Enhancer Based on Singular Spectrum Analysis[J]. IEEE Transactions on Biomedical Engineering, 2012, 59(2): 428-434.
[18] ESTER M, KRIEGEL H P, SANDER J, et al. A Density-based Algorithm for Discovering Clusters in Large Spatial Databases with Noise[C] //Proceedings of the 2nd International Conference on Knowledge Discovery and Data Mining. Portland, Oregon: ACM, 1996: 226-231.
[19] 徐京京.基于聚类和神经网络的异常数据识别算法研究[D].北京:华北电力大学(北京),2019.
[20] IENCO D,BORDOGNA G.Fuzzy Extensions of the DBSCAN Clustering Algorithm[J].Soft Computing,2018,22:1719-1730.
[21] 张 沛.基于神经网络的风电功率短期与超短期预测[D].济南:山东大学,2020.
PDF(1222 KB)
