Anomaly Detection in Dam Behaviour with Machine Learning Classification Models

被引:26
作者
Salazar, Fernando [1 ]
Conde, Andre [1 ]
Irazabal, Joaquin [1 ]
Vicente, David J. [1 ]
机构
[1] Univ Politecn Cataluna, Int Ctr Numer Methods Engn CIMNE, Barcelona 08034, Spain
关键词
anomaly detection; machine learning; support vector machines; random forest; one-class classification; SUPPORT VECTOR MACHINES; RELIABILITY-ANALYSIS; EARTH DAM; RISK;
D O I
10.3390/w13172387
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Dam safety assessment is typically made by comparison between the outcome of some predictive model and measured monitoring data. This is done separately for each response variable, and the results are later interpreted before decision making. In this work, three approaches based on machine learning classifiers are evaluated for the joint analysis of a set of monitoring variables: multi-class, two-class and one-class classification. Support vector machines are applied to all prediction tasks, and random forest is also used for multi-class and two-class. The results show high accuracy for multi-class classification, although the approach has limitations for practical use. The performance in two-class classification is strongly dependent on the features of the anomalies to detect and their similarity to those used for model fitting. The one-class classification model based on support vector machines showed high prediction accuracy, while avoiding the need for correctly selecting and modelling the potential anomalies. A criterion for anomaly detection based on model predictions is defined, which results in a decrease in the misclassification rate. The possibilities and limitations of all three approaches for practical use are discussed.
引用
收藏
页数:22
相关论文
共 45 条
[1]  
Bennett KP, 2000, ACM SIGKDD EXPLORATI, V2, P1, DOI DOI 10.1145/380995.380999
[2]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[3]   Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[4]   A Support Vector Machine Forecasting Model for Typhoon Flood Inundation Mapping and Early Flood Warning Systems [J].
Chang, Ming-Jui ;
Chang, Hsiang-Kuan ;
Chen, Yun-Chun ;
Lin, Gwo-Fong ;
Chen, Peng-An ;
Lai, Jihn-Sung ;
Tan, Yih-Chi .
WATER, 2018, 10 (12)
[5]   Two online dam safety monitoring models based on the process of extracting environmental effect [J].
Cheng, Lin ;
Zheng, Dongjian .
ADVANCES IN ENGINEERING SOFTWARE, 2013, 57 :48-56
[6]   Statistical model optimized random forest regression model for concrete dam deformation monitoring [J].
Dai, Bo ;
Gu, Chongshi ;
Zhao, Erfeng ;
Qin, Xiangnan .
STRUCTURAL CONTROL & HEALTH MONITORING, 2018, 25 (06)
[7]   Gene selection and classification of microarray data using random forest -: art. no. 3 [J].
Díaz-Uriarte, R ;
de Andrés, SA .
BMC BIOINFORMATICS, 2006, 7 (1)
[8]   Urban Flood Mapping Based on Unmanned Aerial Vehicle Remote Sensing and Random Forest Classifier-A Case of Yuyao, China [J].
Feng, Quanlong ;
Liu, Jiantao ;
Gong, Jianhua .
WATER, 2015, 7 (04) :1437-1455
[9]   Anomaly detection in earth dam and levee passive seismic data using support vector machines and automatic feature selection [J].
Fisher, Wendy D. ;
Camp, Tracy K. ;
Krzhizhanovskaya, Valeria V. .
JOURNAL OF COMPUTATIONAL SCIENCE, 2017, 20 :143-153
[10]   Crack Detection in Earth Dam and Levee Passive Seismic Data Using Support Vector Machines [J].
Fisher, Wendy D. ;
Camp, Tracy K. ;
Krzhizhanovskaya, Valeria V. .
INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE 2016 (ICCS 2016), 2016, 80 :577-586