polito.it
Politecnico di Torino (logo)

parallel matlab code for seismic tomography data inversion

Mohammed Eltigani Ali Adam

parallel matlab code for seismic tomography data inversion.

Rel. Alberto Godio. Politecnico di Torino, Corso di laurea magistrale in Petroleum Engineering (Ingegneria Del Petrolio), 2018

[img]
Preview
PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview
[img] Archive (ZIP) (Documenti_allegati) - Altro
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (177MB)
Abstract:

parallel matlab code for seismic tomography data inversion Abstract: The objective of the thesis is to verify the effectiveness of staggered grid mesh applied to two known inversion methods in resolution enhancement. The staggered grid is an approach to enhance resolution and smoothness by averaging several coarse grid inversion solutions by shifting an original grid in both directions in case of 2D inversion without risk of undetermination associated with a fine grid. In this thesis we have taken an overview about the traveltime seismic tomography, with a focus on crosshole tomography and the use of staggered grid for inversion enhancement. The practical work was to improve some properties of an already developed matlab code (SGRAT), which is stand for Staggered Grid Random Tomography invented by Godio Alberto, and Luigi Sambuelli 2011; besides recovering some running errors and an application to synthetic and real field datasets. Some minor running errors have been recovered in forward modelling, conjugated gradient inversion tools regarding the velocity modelling and ray tracing and displaying the results. Plots for spatial coverage (total length per pixel) and ray density (total number of rays per pixel) have been introduced to the simultenious iterative reconustruction technique SIRT. Two data sets have been tested on SIRT inversion first then SIRT staggered random inversion and regular SIRT staggered inversion; results show some kind of resolution improvement measured by root mean square of data errors and velocity map image smoothness. Some open issues recommended to be investigated are the reliability of the conjugated gradient inversion method, effect of parallel computing on reducing computation time.

Relatori: Alberto Godio
Anno accademico: 2018/19
Tipo di pubblicazione: Elettronica
Numero di pagine: 74
Soggetti:
Corso di laurea: Corso di laurea magistrale in Petroleum Engineering (Ingegneria Del Petrolio)
Classe di laurea: Nuovo ordinamento > Laurea magistrale > LM-35 - INGEGNERIA PER L'AMBIENTE E IL TERRITORIO
Aziende collaboratrici: NON SPECIFICATO
URI: http://webthesis.biblio.polito.it/id/eprint/9907
Modifica (riservato agli operatori) Modifica (riservato agli operatori)