Ant Colony Optimisation Inversion of Surface and Borehole Magnetic Data under Lithological Constraints

Shuang Liu,Xiangyun Hu,Tianyou Liu,Yufei Xi,Jianchao Cai,Henglei Zhang

DOI: https://doi.org/10.1016/j.jappgeo.2014.11.010

IF: 1.845

2015-01-01

Journal of Applied Geophysics

Abstract:The ant colony optimisation algorithm has successfully been used to invert for surface magnetic data. However, the resolution of the distributions of the recovered physical property for deeply buried magnetic sources is not generally very high because of geophysical ambiguities. We use three approaches to deal with this problem. First, the observed surface magnetic data are taken together with the three-component borehole magnetic anomalies to recover the distributions of the physical properties. This cooperative inversion strategy improves the resolution of the inversion results in the vertical direction. Additionally, as the ant colony tours the discrete nodes, we force it to visit the nodes with physical properties that agree with the drilled lithologies. These lithological constraints reduce the non-uniqueness of the inversion problem. Finally, we also implement a K-means cluster analysis for the distributions of the magnetic cells after each iteration, in order to separate the distributions of magnetisation intensity instead of concentrating the distribution in a single area. We tested our method using synthetic data and found that all tests returned favourable results. In the case study of the Mengku iron-ore deposit in northwest China, the recovered distributions of magnetisation are in good agreement with the locations and shapes of the magnetite orebodies as inferred by drillholes. Uncertainty analysis shows that the ant colony algorithm is robust in the presence of noise and that the proposed approaches significantly improve the quality of the inversion results.