New Inversion Approach for Interpreting Gravity Data Caused by Dipping Faults

Abstract

Abstract A new inversion algorithm of gravity data utilizing the particle swarm optimization (PSO) algorithm is used to interpret dipping faults models. The PSO technique is stochastic in nature; its development was motivated by the common in‐flight performance of birds looking for food. Particles or models represent birds. Individual particles have a location and a velocity vector. The parameter value represents the position vectors. PSO is adjusted by random particles or models and searches for targets by updating generations. This algorithm determines the dipping faults different parameters (amplitude factor, depth to the center of the upper part of the layer, depth to the center of the lower part of the layer, fault dip angle, and the origin of the fault trace). Herein, the PSO algorithm is applied to noise‐free synthetic data, synthetic data contaminated with different random noise levels (5%, 10%, and 15%) and real field gravity data from Egypt. The applicability and efficiency of the PSO inversion algorithm are well demonstrated for synthetic and field gravity data. The errors of the different estimated parameters are calculated for synthetic data, also, the root mean square error is calculated for synthetic and real data. The parameters estimated from real data matches well with that resulted from different published techniques. From the results obtained by using the present technique, we can apply the proposed technique in different applications, like mining and ore exploration.

Keywords

• depth
• dipping fault
• PSO
• RMS