Application of a modified wild horse optimizer in the inversion of Cole-Cole model parameters

In this study, a modified wild horse optimizer (MWHO) algorithm characterized by a strong evolutionary capability, a faster search speed, and an enhanced optimization ability was introduced into the inversion of Cole-Cole model parameters using the time-domain induced polarization method. The performance of the MWHO algorithm was verified using known models, followed by a comparative study on the derived model parameters. In the case where the time window number and time delay were set at 8 or above and 100~250 ms, respectively, Cole-Cole model parameters with a maximum error of 0.03% were achieved through inversion using the MWHO algorithm. Furthermore, this algorithm exhibited a rapid computation speed and high result stability. Further verification using measured data indicates high consistency among the inverted values of frequency correlation coefficient (c), chargeability (m), and time constant (τ). The inversion results of measured data exhibited low-amplitude frequency correlation coefficient anomalies, high-amplitude chargeability anomalies, and high-amplitude time constant anomalies. These characteristics suggest the presence of metal sulfide ore (mineralized) bodies with a high mineralization degree, high metallic mineral contents, uneven mineralized particles, and high interconnectivity of conductive minerals. Drilling results reveal that siltstones with pyrite alteration occur at depths ranging from 45.08 m to 45.88 m. Disseminated and veinlet pyrite are continuously distributed along fractures. Furthermore, gold mineralization is visible, with gold grade reaching the industrial grade. The inversion results are roughly consistent with the drilling results. Therefore, the MWHO algorithm can be effectively applied to the inversion of Cole-Cole model parameters. Furthermore, this algorithm can distinguish weak induced polarization anomalies, thereby significantly enhancing the resolution of geobodies and highlighting ore-related anomalies. The proposed MWHO algorithm delivers high performance in identifying metal sulfide anomalies, warranting further research and widespread application.