Fuzzy logic algorithms in the analysis of electrotelluric data with reference to monitoring of volcanic activity

The expert processing of monitoring data of large networks on hazardous natural phenomena becomes increasingly more complicated due to an increase in the initial data flow. An approach alternative to the visual recognition of signals is proposed. A number of recognition algorithms and results of the...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Izvestiya. Physics of the solid earth Ročník 43; číslo 7; s. 597
Hlavní autoři: Bogoutdinov, Sh R, Agayan, S M, Gvishiani, A D, Graeva, E M, Rodkin, M V, Zlotnicki, J, Le Mouél, J L
Médium: Journal Article
Jazyk:angličtina
Vydáno: Dordrecht Springer Nature B.V 01.07.2007
Témata:
Algorithms
Monitoring systems
Precipitation
Volcanic eruptions
Volcanoes
ISSN:1069-3513, 1555-6506
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The expert processing of monitoring data of large networks on hazardous natural phenomena becomes increasingly more complicated due to an increase in the initial data flow. An approach alternative to the visual recognition of signals is proposed. A number of recognition algorithms and results of their application to the analysis of geoelectric potential monitoring data are discussed. Data of monitoring La Fournaise Volcano (Réunion Island) obtained in the vicinity of the intense volcanic eruption of 1988 are used. The obtained results show that these algorithms are capable of recognizing anomalous segments of records and discriminating between several types of anomalies presumably associated with the effects of various physical factors (heavy atmospheric precipitation, hydrothermal processes, and so on). The algorithms proposed in this work can be used both for the automation of expert work in operating monitoring systems and in investigations aimed at the identification of typical morphologic sequences in time series of data of various origins.[PUBLICATION ABSTRACT]
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:1069-3513
1555-6506
DOI:10.1134/S1069351307070099