As a non-contact method, the transient electromagnetic (TEM) method has the characteristics of high efficiency, small impact of device, no limitation of site range, and high resolution, and is a hot topic in current research. However, the research on the refined data processing method of TEM is lag, which seriously restricts the application in superficial engineering investigation and is a key problem that needs to be solved urgently. The particle swarm optimization (PSO) algorithm and firefly algorithm (FA) were successful swarm intelligence algorithms inspired by nature. However, the accuracy and efficiency of the algorithm restrict its further development. In this paper, the particle moving velocity of FA algorithm is defined according to the concept of particle moving velocity in PSO algorithm, so as to improve the local fast convergence ability of FA algorithm. On this basis, the appropriate velocity of particle movement is improved, so that the improved algorithm can overcome the oscillation problem around the optimal solution and improve the computational efficiency. And finally, an improved PSO-IFA hybrid optimization algorithm (PSO-IFAH) was proposed in the paper. The proposed algorithm can exploit the strong points of both PSO and FA algorithm mechanisms. A typical layered model was established, and the PSO algorithm, FA algorithm, and PSO-IFAH algorithm were applied to inversion calculations. The results show that the PSO-IFAH algorithm improves calculation accuracy by more than 80% and efficiency by over 60% compared to the PSO and FA algorithms, respectively. The PSO-IFAH algorithm also exhibits high inversion accuracy and stability, with superior anti-noise properties compared to the other algorithms. When implemented in ground TEM measurement data processing, the PSO-IFAH algorithm enhances the resolution of anomalies and low-resistance details, aligning well with actual excavation results. This highlights the algorithm's capability to depict underground electrical structures and karst developments accurately, thereby improving the precision of TEM data processing and interpretation.
Copyright: © 2025 Xu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.