Clustering of lightning discharges
https://doi.org/10.25587/SVFU.2019.69.25526
Abstract
The paper discusses cluster analysis methods for identifying and finding the main parameters of a thunderstorms. Unlike many statistical procedures, cluster analysis methods are used in most cases when there are no a priori hypotheses about classes and the researcher is in the descriptive stage of the study. To solve the problem, it is proposed to use successively two methods of cluster analysis using the Mahalanobis metric and the normalized еuclidean metric. The results obtained using the first method do not contradict the data used in modern theory and practice, but, nevertheless, require a certain accuracy in determining the shape of a cluster. This is due to the fact that, on the one hand, as a result of the use of certain algorithms, “fibrous” clusters are created, in other words, clusters of elliptical shape. On the other hand, thunderstorm clouds in the shape of an ellipse do appear on the fronts, called frontal thunderstorms. In this case, the shape of the hearth is determined by the air flow, pushing against a thunderstorm cloud on the leeward side, giving it an oval shape. The use of the second method makes it possible to exclude noise in the construction of maps of densities of lightning discharges and to improve the accuracy of determining parameters during lightning registration by passive radio engineering methods.
About the Author
S. N. Shabaganova
M.K. Ammosov North-Eastern Federal University
Russian Federation
Russian Federation
Shabaganova Svetlana Nikolaevna – Senior Lecturer Departament of Mining, Mirny Polytechnic Institute
Yakutsk
References
1. Dyuran B. Klasternyj analiz / B. Dyuran, P. Odell. – M.: Statistika, 1977. – 128 s.
2. Leonchik E. YU. Klasternyj analiz. Terminologiya, metody, zadachi / E. YU. Leonchik, O. V. Savastru. – Odessa: ONU im. I.I. Mechnikova, 2007. – 67 s.
3. Byuyul' A. SPSS: iskusstvo obrabotki informacii. Analiz statisticheskih dannyh i vosstanovlenie skrytyh zakonomernostej / A. Byuyul', P. Cefel', V.E. Momot. – SPb.: OOO «DivSoft YUP», 2002. – 601 s.
4. Faktornyj, diskriminantnyj i klasternyj analiz / O. Dzh. Kim [i dr.]; pod red. I.S. Enyukova. – M.: Finansy i statistika, 1989. – 215 s.
5. Petruk V. Primenenie lokal'nyh binarnyh shablonov k resheniyu zadachi raspoznavaniya lic / V. Petruk, A. V. Samorodov, I. N. Spiridonov // Vestnik Moskovskogo gosudarstvennogo tekhnicheskogo universiteta im. N.EH. Baumana. Seriya «Priborostroenie». – 2011. – №. 5. – S. 58-63.
6. SHitikov V. K. Kolichestvennaya gidroehkologiya: metody sistemnoj identifikacii / V.K. SHitikov, G.S. Rozenberg, T.D. Zinchenko. – Tol'yatti: IEHVB RAN, 2003. – 463 s.
7. Mahalanobis P. C. Analysis of race-mixture in Bengal. Presidential Address. – Indian Science Congress, 1925.
8. Hant EH. Iskusstvennyj intellekt / EH. Hant. – M.: Mir, 1978. – 558 s.
9. Duda R. Raspoznavanie obrazov i analiz scen / R. Duda, R. Hart. – M.: Mir, 1976. – 509 s.
10. Steinhaus H. Sur la division des corp materiels en parties / H. Steinhaus // Bull. Acad. Polon. Sci. – 1956. – T. 1, №. 804. – S. 801.
11. Lloyd S. P. Least squares quantization in PCM / S. P. Lloyd // IEEE Transactions on information theory. – 1982. – Vol. IT-28, №2. – S. 129-137.
12. Aleksandrov M. S. Issledovanie atmosfernyh radiopomekh ONCH- i NCH diapazonov i ih istochnikov / M. S. Aleksandrov // Uspekhi sovremennoj radioehlektroniki. – 1998. – №. 10. – S. 3-25.
13. Kononov I. I. Klasternyj analiz grozovoj aktivnosti / I.I. Kononov, I.E. YUsupov // Radiotekhnika i ehlektronika. – 2004. – T. 49, №. 3. – S. 283-291.
14. Zagorujko N. G. Algoritmy obnaruzheniya ehmpiricheskih zakonomernostej / N.G. Zagorujko, V.N. Elkina, G.S. Lbov. – Novosibirsk: Nauka, 1985. – 110 s.
15. Vasil'ev A. E. Grozovaya aktivnost' na Vostoke Sibiri po nablyudeniyam s pomoshch'yu odnopunktovogo grozopelengatora-dal'nomera: dis. … kand. f.-m. nauk: 25.29.00 – Fizika atmosfery i gidrosfery / Institut kosmofizicheskih issledovanij i aehronomii im. YU.G. SHafera SO RAN. YAkutsk, 2005. 147 s.
16. Adzhieva A. A. Klasternyj analiz v avtomaticheskom vyyavlenii i soprovozhdenii grozovyh ochagov po dannym grozopelengacionnoj seti [EHlektronnyj resurs] / A.A. Adzhieva, V.A. SHapovalov // Inzhenernyj vestnik Dona: ehlektr. nauch. zhurn. – 2016. – №2. – Rezhim dostupa: http://ivdon.ru. – (Data obrashcheniya: 21.12.2016).
17. Harakteristiki grozovyh yacheek po nablyudeniyam v YAkutii / SHabaganova S.N. [i dr.] // Meteorologiya i gidrologiya. – 2012. – № 12. – S. 35–43.
18. Mandel' I. D. Klasternyj analiz / I.D. Mandel'. – M.: Finansy i statistika, 1988. – 176 s.
19. Wishart D. Mode analysis: A generalization of nearest neighbor which reduces chaining effects / D. Wishart // Numerical taxonomy. – 1969. – T. 76, №. 282-311. – S. 17.
20. Panyukov A. V. Sistemy passivnogo monitoringa grozovoj deyatel'nosti / A.V. Panyukov, D.V. Buduev, D.N. Malov // Vestnik YUzhno-Ural'skogo gosudarstvennogo universiteta. Seriya: Matematika. Fizika. Himiya. – 2003. – №. 8. – S. 24.
21. Kohn M. et al. Nowcasting thunderstorms in the Mediterranean region using lightning data / M. Kohn [et al.] // Atmospheric Research. – 2011. – T. 100, №. 4. – S. 489-502.
Review
For citations:
Shabaganova S.N. Clustering of lightning discharges. Vestnik of North-Eastern Federal University. 2019;(1):58-69. (In Russ.) https://doi.org/10.25587/SVFU.2019.69.25526
Views:
158
Email this article 