Inversion of magnetic and gravity anomaly using artificial neural networks

Abstract

In this study, the Artificial Neural Network Regression technique has been applied for inversion of gravity and magnetic anomaly datasets. Using this technique, we have tried to find the unknown parameters of the causative source, i.e., density contrast, the radius of the buried structure, amplitude coefficient, angle of effective magnetization, depth of the source, and horizontal position. The algorithm implemented by constraining shape factor for better results of other parameters. The applicability of this algorithm is tested both for noise-free and 10% gaussian noisy synthetic gravity and magnetic anomaly data generated for simple geometrical bodies like a sphere, horizontal cylinder and thin dyke.