Novelty detection using extreme value statistics

The first and most important objective of any damage identification algorithm is to ascertain with confidence if damage is present or not. Many methods have been proposed for damage detection based on ideas of novelty detection founded in pattern recognition and multivariate statistics. The philosophy of novelty detection is simple. Features are first extracted from a baseline system to be monitored, and subsequent data are then compared to see if the new features are outliers, which significantly depart from the rest of population. In damage diagnosis problems, the assumption is that outliers are generated from a damaged condition of the monitored system. This damage classification necessitates the establishment of a decision boundary. Choosing this threshold value is often based on the assumption that the parent distribution of data is Gaussian in nature. While the problem of novelty detection focuses attention on the outlier or extreme values of the data, i.e., those points in the tails of the distribution, the threshold selection using the normality assumption weights the central population of data. Therefore, this normality assumption might impose potentially misleading behavior on damage classification, and is likely to lead the damage diagnosis astray. In this paper, extreme value statistics is integrated with the novelty detection to specifically model the tails of the distribution of interest. Finally, the proposed technique is demonstrated on simulated numerical data and time series data measured from an eight degree-of-freedom spring-mass system.

Download Full-text

Assessing Regional Scale Variability in Extreme Value Statistics Under Altered Climate Scenarios

10.2172/1170412 ◽

2015 ◽

Author(s):

Nathaniel Brunsell ◽

David Mechem ◽

Chunsheng Ma

Keyword(s):

Regional Scale ◽

Extreme Value ◽

Extreme Value Statistics ◽

Climate Scenarios

Download Full-text

On the spectral shape of SEP events: An extreme value statistics approach

10.1063/1.4811027 ◽

2013 ◽

Author(s):

M. Laurenza ◽

G. Consolini ◽

M. Storini ◽

A. Damiani

Keyword(s):

Extreme Value ◽

Spectral Shape ◽

Extreme Value Statistics

Download Full-text

Extreme value statistics and return intervals in long-range correlated uniform deviates

Physical Review E ◽

10.1103/physreve.79.041131 ◽

2009 ◽

Vol 79 (4) ◽

Cited By ~ 21

Author(s):

N. R. Moloney ◽

J. Davidsen

Keyword(s):

Long Range ◽

Extreme Value ◽

Extreme Value Statistics

Download Full-text

Extreme value statistics of mutation accumulation in renewing cell populations

Physical Review E ◽

10.1103/physreve.98.050401 ◽

2018 ◽

Vol 98 (5) ◽

Cited By ~ 1

Author(s):

Philip Greulich ◽

Benjamin D. Simons

Keyword(s):

Mutation Accumulation ◽

Extreme Value ◽

Extreme Value Statistics ◽

Cell Populations

Download Full-text

Bestimmung seltener Starkniederschläge kurzer Dauer - Fallbeispiele im Vergleich mit den schweizerischen Starkniederschlagskarten | Determination of Rare Rainfall Intensities of Short-Duration - Case Studies Compared with Maps of Intense Rainfall in Switzerland

Schweizerische Zeitschrift fur Forstwesen ◽

10.3188/szf.1999.0209 ◽

1999 ◽

Vol 150 (6) ◽

pp. 209-218 ◽

Cited By ~ 2

Author(s):

Felix Forster ◽

Walter Baumgartner

Keyword(s):

Case Studies ◽

Short Duration ◽

Extreme Value ◽

Extreme Value Statistics ◽

Intense Rainfall

The two maps of intense rainfall in the Hydrological Atlas of Switzerland (1992, 1997) are compared to data of an evaluation of extreme value statistics. The results are transferred to recommendations for practioners.

Download Full-text