Local geoid modeling in Azerbaijan using ANN، ANFIS Methods
Subject Areas : journal of Artificial Intelligence in Electrical Engineering
1 -
Abstract :
In order to measure the height in mapping engineering topics, we need to define the height base level or geoid. In fact, the geoid is one of the earth's equipotential surfaces, which best approximates the mean sea level (MSL) based on the least squares method. The separation between the planar base surface (elliptical) and the geoid is called the height of the geoid (N). By having this quantity, orthometric height (Ho) and normal height (h) can be converted to each other. There are different methods to determine the height of the geoid. In this thesis, the efficiency of machine learning models to determine geoid height locally and using GPS/Leveling measurements is evaluated. In order to do this, the geodetic coordinates of 26 stations of the northwestern network of Iran, whose orthometric height was also measured by the first level alignment by the National Mapping Organization (NCC), were used. In these stations, the orthometric height difference from normal height (h) is considered as geoid height (N). Therefore, the input of ANN, ANFIS models is the geodetic latitude and longitude coordinates of the stations and the corresponding output is the geoid height. The models have been trained using 22 and 19 stations. . Comparison of RMSE shows that ANN model with less number of training stations provides higher accuracy than ANFIS models. The results of this paper show that using ANN and ANFIS models, geoid height can be locally estimated and used with high accuracy.
Ardalan, A.A., Hatam, C.Y., Sharifi, M.A., Safari, A., Gazavi, K., Motagh M., (2002). Determination of Precise geoid for Iran based on Potential approach. Technical report, National Cartographic Center of Iran (NCC), Tehran, Iran.
Kiamehr, R., (2006). Precise Gravimetric Geoid Model for Iran Based on GRACE and SRTM Data and the Least-Squares Modification of Stokes’ Formula: with Some Geodynamic Interpretations. PhD Thesis. Royal Institute of Technology, Stockholm, Sweden.
Kavzoglu, T., Saka, M. (2005). Modeling local GPS/leveling geoid undulations using artificial neural networks. J Geodesy 78, 520–527, doi.org/10.1007/s00190-004-0420-3.
Mars, P., J.R. Chen, R. Nambiar (1996). Learning Algorithms: Theory and Applications in Signal Processing, Control and Communications, CRC Press, Boca Raton, Florida.
Najafi M., (2004). Determination of Precise Geoid for Iran Based on Stokes-Helmert Scheme. Report 2003. TOTAK Project, National Cartographic Center of Iran (NCC), Tehran, Iran.
Zaletnyik, P., Völgyesi, L., Paláncz, B., (2007). Modeling local GPS/leveling geoid undulations using Support Vector Machines. Civil Engineering 52/1 (2008) 39–43, doi: 10.3311/pp.ci.2008-1.06.
