Application of Unsupervised Anomaly Detection Techniques to Moisture Content Data from Wood Constructions

被引：6

作者：

Faura, Alvaro Garcia ^{[1
]}

Gtepec, Dejan ^{[1
]}

Cankar, Matija ^{[1
]}

Humar, Miha ^{[2
]}

机构：

[1] XLAB Doo, Pot Brdom 100, SI-1000 Ljubljana, Slovenia

[2] Univ Ljubljana, Biotech Fac, Dept Wood Sci & Technol, SI-1000 Ljubljana, Slovenia

来源：

FORESTS | 2021年 / 12卷 / 02期

基金：

欧盟地平线“2020”;

关键词：

wood moisture monitoring; Unsupervised Anomaly Detection (UAD); Moisture Content (MC) data; wooden facade and windows; PERFORMANCE; PRODUCTS; DECAY;

D O I：

10.3390/f12020194

中图分类号：

S7 [林业];

学科分类号：

0829 ; 0907 ;

摘要：

Wood is considered one of the most important construction materials, as well as a natural material prone to degradation, with fungi being the main reason for wood failure in a temperate climate. Visual inspection of wood or other approaches for monitoring are time-consuming, and the incipient stages of decay are not always visible. Thus, visual decay detection and such manual monitoring could be replaced by automated real-time monitoring systems. The capabilities of such systems can range from simple monitoring, periodically reporting data, to the automatic detection of anomalous measurements that may happen due to various environmental or technical reasons. In this paper, we explore the application of Unsupervised Anomaly Detection (UAD) techniques to wood Moisture Content (MC) data. Specifically, data were obtained from a wood construction that was monitored for four years using sensors at different positions. Our experimental results prove the validity of these techniques to detect both artificial and real anomalies in MC signals, encouraging further research to enable their deployment in real use cases.

引用

页码：1 / 19

页数：19

共 43 条

[11]

Cleveland R. B., 1990, J. Off. Stat, V6, P3, DOI DOI 10.1007/978-1-4613-4499-5_24

[12] Anomaly Detection for IoT Time-Series Data: A Survey [J].

Cook, Andrew A. ;

Misirli, Goksel ;

Fan, Zhong .

IEEE INTERNET OF THINGS JOURNAL, 2020, 7 (07) :6481-6494

[13]

Farrar C.R., 2012, Structural Health Monitoring: A Machine Learning Perspective

[14] Multi-level Anomaly Detection in Industrial Control Systems via Package Signatures and LSTM networks [J].

Feng, Cheng ;

Li, Tingting ;

Chana, Deeph .

2017 47TH ANNUAL IEEE/IFIP INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS (DSN), 2017, :261-272

[15]

Goasduff L., INTERNET THINGS MARK

[16] ANN-based soft-sensor for real-time process monitoring and control of an industrial polymerization process [J].

Gonzaga, J. C. B. ;

Meleiro, L. A. C. ;

Kiang, C. ;

Maciel Filho, R. .

COMPUTERS & CHEMICAL ENGINEERING, 2009, 33 (01) :43-49

[17] Discovering cluster-based local outliers [J].

He, ZY ;

Xu, XF ;

Deng, SC .

PATTERN RECOGNITION LETTERS, 2003, 24 (9-10) :1641-1650

[18] Long short-term memory [J].

Hochreiter, S ;

Schmidhuber, J .

NEURAL COMPUTATION, 1997, 9 (08) :1735-1780

[19] Moisture Performance of Facade Elements Made of Thermally Modified Norway Spruce Wood [J].

Humar, Miha ;

Lesar, Bostjan ;

Krzisnik, Davor .

FORESTS, 2020, 11 (03)

[20] The Performance of Wood Decking after Five Years of Exposure: Verification of the Combined Effect of Wetting Ability and Durability [J].

Humar, Miha ;

Krzisnik, Davor ;

Lesar, Bostjan ;

Brischke, Christian .

FORESTS, 2019, 10 (10)

← 1 2 3 4 5 →