COMPARISON OF VERY FAST SIMULATED ANNEALING (VFSA) AND MODIFIED PARTICLE SWARM OPTIMIZATION (MPSO) INVERSION METHOD FOR 1D TDEM DATA MODELLING

• Main Author: SUDHA PRABAWA NIM: 12314013, RAI
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/30168
• Institution: Institut Teknologi Bandung

<p align="justify">Time domain electromagnetic method (TDEM) is a geophysical method method that can be applied to a wide range of problem. TDEM data requires non-linear inversion in its modelling because the relationship between the data and model is complex. The global approach is an approach in nonlinear inversion calculation which is not sensitive to the initial model and can obtain the global minimum. Previous studies proved that one kind of inversion method will not result in the same performance for every type of problems. For this reason, this study was conducted to discuss the details of the performance of global non-linear inversion methods so that the TDEM inversion process can be performed more optimally. This study compares the performance of VFSA and MPSO in inversion modelling of synthetic and field 1D TDEM data. The inversion modelling is done multiple times to consider the random factor present in both methods and also to compare the result’s variability of each method. From the result of synthetic data inversion, it can be seen that there is an ambiguity in 1D TDEM which causes a variation of inversion performance based on the model. Based on the comparison of the two merhods, the MPSO works better in inversion of data with lower level of ambiguity and vice versa. Overall, it is concluded that the VFSA method performs better than MPSO because it has lower variabilty and gives better inversion results in 1D TDEM modelling which has formidable ambiguity.<p align="justify">