Approximating the Performance of a Time-Domain Pulsed Induction EMI Sensor with Multiple Frequency-Domain FEM...

Area of Science:

• Geophysics
• Electromagnetism
• Sensor Technology

Background:

• Developing Pulsed Induction (PI) electromagnetic induction (EMI) sensors for Arctic applications is hindered by environmental inaccessibility and high costs of in situ testing.
• Robust simulation strategies are crucial for optimizing PI sensor features like coil turns and geometry.
• Previous work established a time-domain finite element model (FEM) for simulating Arctic PI sensors.

Purpose of the Study:

• To present a novel method for approximating time-domain PI sensor simulations using multiple frequency-domain simulations.
• To validate this frequency-domain approach against empirical data.
• To demonstrate the method's capability in enabling broader sensor performance comparisons over a wider parameter space.

Main Methods:

• Approximation of time-domain PI sensor simulations using multiple frequency-domain simulations.
• Comparison of the Fast Fourier Transform (FFT) of time-domain simulations with collections of frequency-domain simulations.
• Validation against empirical data using a PI sensor over seawater with an air gap simulating sea ice.

Main Results:

• The frequency-domain simulation method was validated against empirical data.
• Simulations covered a range of coil dimensions (0.5x0.5 m to 1.0x2.0 m).
• A significant reduction in simulation time was achieved, from 38,013 minutes for a time-domain simulation to 141 minutes for a frequency-domain simulation (for 10 lift-off distances).

Conclusions:

• The proposed frequency-domain simulation method offers a computationally efficient alternative to time-domain simulations for PI sensor development.
• This approach facilitates the optimization and comparison of sensor designs over an expanded parameter space.
• The method effectively streamlines the development of PI sensors for challenging environments like the Arctic.