Depth Estimation Of Salt Domes Using Gravity Data Through MLP Neural Network, Case Study: Mors Salt Dome in Denmark

Artificial neural network analysis (ANN) is a new tool, which mimics the human's brain. Today, this tool has a variety of applications in science, engineering, social science, economic and etc. Along with these applications, it potentially has wide usage in the oil industry and geophysics as well. In this paper, the method of Artificial Neural Networks is used as a suitable tool for intelligent interpretation of gravity data. We aim to model geological features known as salt domes in order to determine the principal diagnostic features of the anomalous body from gravity data, firstly by 2D forward modeling then with a Multi-Layer Perceptron (MLP) which has been trained for depth estimation of salt domes from gravity data. The approach was applied to both synthetic and real data. The real data is gravity data collected over the Mors Salt dome located in Denmark and the estimated depth from these methods is very close to the real depth of the salt dome. However, MLP is shown to be faster than other neural networks with less data required for the training.