Implementation of Bat Algorithm (BA) in Vertical Electrical Sounding (VES) Data Inversion

Bat Algorithm (BA) is a population-based metaheuristic algorithm inspired by the echolocation mechanism of bats. This algorithm has great potential in solving nonlinear optimization problems, including those encountered during the process of vertical electrical sounding (VES) data inversion for subsurface exploration. This study aims to test the effectiveness and validity of BA in inverting VES data through parameter testing and implementation on synthetic data and field data. The method employed involves tuning the BA parameter to determine the optimal configuration and accuracy testing using both noise-free synthetic data and data with 10% noise added. The results show that the optimal parameter configuration is obtained at an initial loudness value of r_0 = 0.01, a loudness reduction coefficient of α = 0.9, and a pulse emission frequency increase coefficient of γ =0.9. The synthetic data inversion produces a high similarity index (〖SI〗_x) value (above 90%) and low misfit (below 7%), indicating accurate and stable BA performance. The implementation of the BA on field data resulted in low inversion errors, consistently below 4%. Application of the inversion to field data acquired in a residential area in South Lampung successfully identified two aquifer categories: an unconfined aquifer at depths of up to 9 meters and a shallow confined aquifer at depths of 7–43 meters, with tuffaceous sandstone identified as the primary candidate aquifer. This study confirms the effectiveness of BA in interpreting VES data for hydrogeological purposes.