Increase of the Resistance to Noise in Data for Neural Network Solution of the Inverse Problem of Magnetotellurics with Group Determination of Parameters

2016 ◽  
pp. 502-509 ◽  
Author(s):  
Igor Isaev ◽  
Eugeny Obornev ◽  
Ivan Obornev ◽  
Mikhail Shimelevich ◽  
Sergey Dolenko
Keyword(s):  
Neural Network ◽  
Inverse Problem ◽  
Network Solution ◽  
Determination Of Parameters

An analog neural network solution to the inverse problem of 'early taction'

1992 ◽  
Vol 8 (2) ◽  
pp. 196-212 ◽  
Author(s):  
Y.C. Pati ◽  
P.S. Krishnaprasad ◽  
M.C. Peckerar
Keyword(s):  
Neural Network ◽  
Inverse Problem ◽  
Network Solution ◽  
Analog Neural Network

scholarly journals Estimation of Accuracy of Determination of Parameters of the Buried Archaeological Objects According to the Data of Different-Altitude Magnetic Survey on the Basis of Numerical Modeling

2020 ◽  
Vol 18 (1) ◽  
pp. 16-26
Author(s):  
N. D. Evmenov ◽  
V. V. Voronin ◽  
P. G. Dyadkov ◽  
O. A. Pozdnyakova
Keyword(s):  
Inverse Problem ◽  
Real Data ◽  
Archaeological Site ◽  
The Body ◽  
Average Error ◽  
Magnetic Survey ◽  
Synthetic Model ◽  
Determination Of Parameters ◽  
Archaeological Objects

An algorithm for solving the inverse problem for determining the parameters of a buried archaeological object in the form of an elliptical cylinder (as a model of a typical burial or pit) is proposed. The estimation of accuracy of determination of parameters of the buried archaeological objects on the basis of the numerical solution of the inverse problem of magnetic prospecting for synthetic model is carried out. Its implementation included, firstly, the stage of modeling the magnetic field from a synthetic model of the archaeological site at three different heights for different parameters of the grid density and the size of the magnetic survey site and, secondly, the stage of solving the inverse problem from synthetic data at different heights, for different parameters of the magnetic survey and assessing the accuracy of the solution of the inverse problem. It is found that the accuracy of determining the parameters of the anomalous body is more affected by the size of the survey area than the density of the measurement network. There is also a tendency to increase the accuracy of determining parameters when using data of different heights. Approbation of the algorithm for solving the inverse problem on real data of different-altitude magnetic survey over the burial at the archaeological site Tartas-1 showed quite good results. The average error in determining the geometric parameters of the body and its occurrence is ~ 15 %.

Sign in / Sign up

Export Citation Format

Share Document