Intelligent algorithm of extreme gradient boosting based perfectly matched monolayer and its efficient absorption on airborne transient electromagnetics problems

Abstract

In addition to requiring accuracy and computational efficiency for solving low-frequency subsurface sensing problem on the airborne transient electromagnetics (ATEMs), to the best of our knowledge, the complexity of subsurface sensing problems should also be considered to decrease more and more computational resources, especially for a large-scale complicated multis-cale problem with the difference between background and targets. For simulating the open-domain, the finite-thickness perfectly matched layer (PML) is used to truncate the computational region, and whereas the whole domain becomes larger so that the problem complexity gets increased. As a result, we propose a novel perfectly matched monolayer (PMM) model based on the eXtreme Gradient Boosting (XGB), which is selected and added to further improve the performance during the finite-difference time-domain (FDTD) simulation. The proposed XGB-based PMM model could achieve higher accuracy by the ensemble learning method of feature attention, and meanwhile has less memory and time consumption. Besides, this model has significant advantages in terms of model training stability and its lightweight due to the fact that it relies on the characteristics of traditional machine learning (ML) models. Finally, three-dimensional numerical simulations of ATEM problems have been carried out to prove the validity and stability of the proposal. The proposed model could not only achieve advantages in numerical accuracy, efficiency and problem complexity, but be integrated into the FDTD solver to process the low-frequency ATEM problems.