A novel method based on improved SFLA for IP information extraction from TEM signals.

Area of Science:

• Geophysics
• Electrical Engineering

Background:

• Transient electromagnetic (TEM) signal analysis is vital for deep resource exploration.
• Extracting induced polarization (IP) information from TEM data is challenging due to limitations in linear inversion methods, including initial condition dependence, local optima, and non-uniqueness.

Purpose of the Study:

• To develop an advanced algorithm for accurate IP information extraction from TEM signals.
• To overcome the limitations of traditional linear inversion techniques in geophysical exploration.

Main Methods:

• An improved shuffle frog leaping algorithm (ISFLA) was developed, integrating tent chaotic distribution and an adaptive mobile factor.
• Tent chaotic distribution enhances initial population distribution for improved global search capability.
• An adaptive mobile factor replaces random operators to balance local and global searches, increasing accuracy and convergence stability.

Main Results:

• The ISFLA method demonstrated superior performance in transient electromagnetic (TEM) inversion for a 1D layered geoelectric model.
• The algorithm effectively reconstructed geoelectric structures and extracted IP information with enhanced robustness.
• ISFLA achieved higher global search ability and inversion accuracy compared to other heuristic algorithms.

Conclusions:

• The proposed ISFLA is a robust and accurate method for extracting induced polarization (IP) information from transient electromagnetic (TEM) signals.
• This algorithm offers significant advantages over existing methods for deep mineral, oil, and gas resource exploration.