A geomagnetic vector compensation method compatible with nonlinear interferences based on back propagation network...

Area of Science:

• Geophysics
• Sensor Technology
• Data Science

Background:

• Geomagnetic vector measurement is crucial for various applications.
• Traditional methods struggle with nonlinear magnetic interferences, limiting accuracy.
• Existing compensation models fail to fully address these complex interferences.

Purpose of the Study:

• To propose a new compensation method for geomagnetic vector measurements.
• To address the limitations of linear compensation models in handling nonlinear interferences.
• To improve the accuracy and reliability of geomagnetic measurements.

Main Methods:

• A back propagation neural network (BPNN) was developed for nonlinear magnetic interference compensation.
• A 3D Helmholtz coil was utilized to generate comprehensive datasets for network training.
• Simulations and experimental validations were performed to assess the method's efficacy.

Main Results:

• The BPNN method demonstrated superior nonlinear mapping capabilities compared to linear models.
• The use of a 3D Helmholtz coil provided flexible and practical data generation.
• Experimental results showed a significant reduction in root mean square errors for all measured components.

Conclusions:

• The proposed BPNN-based compensation method effectively mitigates nonlinear magnetic interferences.
• The technique offers a substantial improvement in geomagnetic vector measurement accuracy.
• This advancement has significant implications for applications requiring precise magnetic field detection.