Author/Authors
güllü, mevlüt afyon kocatepe üniversitesi - mühendislik fakültesi - harita mühendisliği bölümü, Turkey , turgut, bayram afyon kocatepe üniversitesi - mühendislik fakültesi - harita mühendisliği bölümü, Turkey , baybura, tamer afyon kocatepe üniversitesi - mühendislik fakültesi - harita mühendisliği bölümü, Turkey
Title Of Article
The Comparison of Artifical Neural Networks and Kriging Interpolation Method fort he Geoid Height Determination
شماره ركورد
27820
Abstract
Global Navigation Satellite System (GNSS) in the determination of vertical datums have been used since 2001 with the spread of satellite technologies in Turkey. The heights given by the GNSS system are ellipsoidal heights. However, Turkish National Vertical Control Network-1999 (TNVCN- 1999) is based on the orthometric height system. Geoid heights are required in order to transform the ellipsoidal heights obtained by GNSS measurements to the orthometric heights. The objective of this study is to determine the geoid heights by Artificial Neural Networks (ANN) and Kriging interpolation method. Kriging method that is one of the interpolation methods widely used and Radial Basis Function Neural Networks (RBFNN) have been compared over the test network consisting of selected points in Afyonkarahisar. The differences between geoid heights calculated by RBFSA and Kriging methods and known geoid heights were evaluated in terms of root mean square error and better results were achieved by RBFNN method.
From Page
674
NaturalLanguageKeyword
Geoid height , Artificial neural networks , RBFNN , Kriging
JournalTitle
Afyon Kocatepe University Journal Of Science and Engineering
To Page
678
JournalTitle
Afyon Kocatepe University Journal Of Science and Engineering
Link To Document