Inversion of Apparent Resistivity Using Regularized Neural Network
R. Shah and S. Kumar
Event name: 80th EAGE Conference and Exhibition 2018
Session: Student Poster: Near Surface Other
Publication date: 11 June 2018
Info: Extended abstract, PDF ( 311.74Kb )
Price: € 20
Since the beginning of the nineteenth century, a significant evolution in optimization theory has been noticed. In the past lots of statistical algorithm have been used to invert apparent resistivity to get layer parameters. But these algorithm are not very stable with very wide range of values. This paper shows the application of a statistically sound algorithm called as neural network which is based on the analogy of human brain. In this paper we have used regularized neural network for the inversion purpose.The inverted model parameters was found to be independent of the search space, thereby showing the robustness of the algorithm.