This record is being processed for inclusion into GeoRef. It may not yet have been indexed, given a translated title, or checked by a GeoRef editor.
Application of back propagation artificial neural networks for gravity field modelling
|Online Access:||Get full text|
|Author Affiliations:||Primary: |
Afyon Kocatepe University, Department of Geomatics, Afyonkarahisar, Turkey
|Volume Title:||Acta Montanistica Slovaca|
|Source:||Acta Montanistica Slovaca, 21(3), p.200-207. Publisher: Fakulta BERG Technickei Univerzity, Kosice, Slovakia. ISSN: 1335-1788|
|Note:||In English. 24 refs.; illus., incl. 3 tables, geol. sketch maps|
|Summary:||Geodesy provides a unique framework for the monitoring, understanding and prognosis of the Earth system as a whole, globally as well as locally. This understanding of geodesy is based on the three pillars: (i) geokinematics, (ii) Earth rotation, and (iii) gravity field. The third pillar of geodesy refers to the knowledge of the geometry of the gravity field of the Earth. The gravity field of the Earth addresses the problems of the transformation of geodetic observations made in physical space (affected by gravity) into geometrical space in which positions are usually defined. In addition, the shapes of equipotential surfaces and plumb lines are needed for projects involving the physical environment (e.g., flow of water). In this paper, the utility of Back Propagation Artificial Neural Network (BPANN) more widely applied in diverse fields of science among all other neural network models is investigated as an alternative tool for gravity field modelling. In order to evaluate the performance of BPANN, the gravity values are also calculated by global geopotential models (EGM2008 and EIGEN-6C4). The results are compared in terms of the root mean square error (RMSE) over a study area. It was concluded that the employment of BPANN can be a feasible gravity calculation tool for the geodetic application.|
|Subjects:||Applications; Artificial intelligence; Case studies; Earth; Geodesy; Geophysical methods; Gravity field; Gravity methods; Interpretation; Mathematical models; Neural networks; Optimization; Plate rotation; Plate tectonics; Propagation; Arizona; California; Nevada; Pacific Coast; United States; Utah|
|Coordinates:||N310000 N310000 W1220000 W1220000|
|Copyright Information:||GeoRef, Copyright 2021 American Geosciences Institute.|
No Tags, Be the first to tag this record!