Parallel Inversion of 1D Magnetotelluric Data Using Particle Swarm Optimization Algorithm

The inversion of magnetotelluric data is a multi-parameter, nonlinear, and multimodal optimization problem. Particle swarm optimization (PSO) algorithm, which was developed by enlightenment of the behavior of birds in looking for food, is the fine solver for this geophysical inversion problem. As the forward problem becomes complex, the inversion becomes much more slow. We proposed a parallel PSO inversion algorithm to speedup the inversion of 1D magnetotelluric data. The numerical results show that the parallel PSO algorithm can speedup the inversion effectively and ensure the solution accuracy.