Design of Fluxgate Current Sensor Based on Magnetization Residence Times and Neural Networks

Jingjie Li,Wei Ren,Yanshou Luo,Xutong Zhang,Xinpeng Liu,Xue Zhang

DOI: https://doi.org/10.3390/s24123752

IF: 3.9

2024-06-10

Sensors

Abstract:This study introduces a novel fluxgate current sensor with a compact, ring-shaped configuration that exhibits improved performance through the integration of magnetization residence times and neural networks. The sensor distinguishes itself with a unique magnetization profile, denoted as M waves, which emerge from the interaction between the target signal and ambient magnetic interference, effectively enhancing interference suppression. These M waves highlight the non-linear coupling between the magnetic field and magnetization residence times. Detection of these residence times is accomplished using full-wave rectification circuits and a Schmitt trigger, with a digital output provided by timing sequence detection. A dual-layer feedforward neural network deciphers the target signal, exploiting this non-linear relationship. The sensor achieves a linearity error of 0.054% within a measurement range of 15 A. When juxtaposed with conventional sensors utilizing the residence-time difference strategy, our sensor reduces linearity error by more than 40-fold and extends the effective measurement range by 150%. Furthermore, it demonstrates a significant decrease in ambient magnetic interference.

engineering, electrical & electronic,chemistry, analytical,instruments & instrumentation

What problem does this paper attempt to address?

### Problems the Paper Attempts to Solve

This paper aims to address the issues of interference from geomagnetic fields and power frequency magnetic fields in traditional current sensors during measurement, and proposes a new current sensor design based on magnetization residence times and neural network technology.

#### Main Issues
1. **Magnetic Field Interference**: Existing current sensors are easily affected by geomagnetic fields and power frequency magnetic fields, leading to inaccurate measurements.
2. **Accuracy and Linearity Error**: Traditional sensors exhibit significant nonlinear errors in weak current detection, limiting their performance in industrial applications.

#### Solutions
1. **New Magnetic Core Structure**: The paper proposes a toroidal magnetic core design that enhances the ability to suppress external interference through unique magnetization waveforms (M waves).
2. **Magnetization Residence Time Reading Strategy**: By analyzing the residence time of the magnetization state, a full-wave rectifier circuit and a Schmitt trigger are used to extract this time information.
3. **Neural Network Processing**: A dual-layer feedforward neural network is introduced to decode the target signal, utilizing the nonlinear relationship between magnetization time and magnetic field to improve detection accuracy.

#### Main Contributions
1. **Magnetization Waveform Explanation**: For the first time, the unique "M" shaped magnetization phenomenon in current sensors is comprehensively explained, and the proposed magnetic micro-element method is used for modeling.
2. **Theoretical Support and Interference Suppression**: By analyzing the nonlinear coupling relationship between magnetization time and magnetic field, a theoretical foundation for current detection and environmental interference suppression is provided.
3. **High-Precision Current Extraction**: Combining neural networks with sensing mechanisms achieves high-precision target current extraction.

In summary, this paper aims to significantly improve the accuracy and anti-interference capability of current sensors by improving the magnetic core structure and introducing neural network technology.