Joint modeling of Rayleigh wave dispersion and H/V spectral ratio using Pareto-based  multiobjective particle swarm optimization

Ersin BÜYÜK; Ekrem ZOR; Abdullah KARAMAN

doi:https://doi.org/10.3906/yer-2001-15

Makale detayları

Joint modeling of Rayleigh wave dispersion and H/V spectral ratio using Pareto-based multiobjective particle swarm optimization

Yazarlar:

Ersin BÜYÜK | Ekrem ZOR | Abdullah KARAMAN

Yayıncı:

TUBITAK

Dergi:

Turkish Journal of Earth Sciences 2020 Vol.29 Issue 4, pp.684-695

Belge Tipi:

Makale

Yayın Tarihi:

4/2020

Belge Dili:

İngilizce

Document ID (Org No.):

2000730834

Sıklık:

İki Aylık

eISSN:

1303-619X

pISSN:

1300-0985

DOI:

https://doi.org/10.3906/yer-2001-15

Özet:

 

Estimating the shear wave velocity and thickness of geologic units in a sedimentary structure is important for quantifying the  local site effect caused by an earthquake. Inversion of the Rayleigh wave dispersion alone is sensitive to the absolute average shear wave  velocity, while the H/V spectral ratio is sensitive to velocity contrasts. The solution of these models in a joint system using conventional  inversion techniques suffers from difficulties while evaluating partial derivatives, dependencies to the initial model that is sometimes  difficult to estimate, and trapping at a local minimum. Herein, a joint model using the Pareto optimality technique with multiobjective  particle swarm optimization was constructed as an effective global optimization method to decrease weakness and increase performance  while estimating the depths and velocities. The presented approach is a multiobjective optimization method that does not require  weighting, and allows evaluations of the individual objective function separately in a joint system. This study presents 2 synthetic joint  system examples to account for a gradient-type and sharp velocity contrast cases, and an application from field data obtained from the  Bursa Basin in Turkey. With these examples, an automated parameter search space was demonstrated to account for the abrupt velocity  changes and a number of additional key points that are essential for a reasonable optimal solution.

Makale detayları

Joint modeling of Rayleigh wave dispersion and H/V spectral ratio using Pareto-based multiobjective particle swarm optimization

Yazarlar: Ersin BÜYÜK | Ekrem ZOR | Abdullah KARAMAN

Yayıncı: TUBITAK

Dergi: Turkish Journal of Earth Sciences 2020 Vol.29 Issue 4, pp.684-695

Belge Tipi: Makale

Yayın Tarihi: 4/2020

Belge Dili: İngilizce

Document ID (Org No.): 2000730834

Sıklık: İki Aylık

eISSN: 1303-619X

pISSN: 1300-0985

DOI: https://doi.org/10.3906/yer-2001-15

Sayı

Yazarlar:

Ersin BÜYÜK | Ekrem ZOR | Abdullah KARAMAN

Yayıncı:

TUBITAK

Dergi:

Turkish Journal of Earth Sciences 2020 Vol.29 Issue 4, pp.684-695

Belge Tipi:

Makale

Yayın Tarihi:

4/2020

Belge Dili:

İngilizce

Document ID (Org No.):

2000730834

Sıklık:

İki Aylık

eISSN:

1303-619X

pISSN:

1300-0985

DOI:

https://doi.org/10.3906/yer-2001-15