TL;DR
This paper introduces a novel multi-stage magnetic ring structure for Tunnel Magnetoresistance (TMR) sensors to improve AC/DC leakage current detection in power equipment. By optimizing the ring's geometry using an advanced genetic algorithm, the sensor achieves significantly enhanced sensitivity and immunity to external magnetic interference, crucial for reliable monitoring in complex environments.
What Problem Does This Paper Address?
The reliable detection of small AC/DC leakage currents in power equipment is critical for ensuring grid stability and preventing faults. Traditional sensors struggle with the low amplitude of these currents and are highly susceptible to external electromagnetic interference, leading to misoperations and inaccurate readings. This is particularly challenging as modern power systems increasingly incorporate nonlinear loads, creating complex current waveforms. While TMR sensors offer high sensitivity, their performance is often compromised by stray magnetic fields. This paper aims to address this by designing a TMR sensor with improved magnetic field concentration and interference rejection capabilities.
Methods
The core innovation lies in the multi-stage magnetic ring structure designed to amplify the magnetic field generated by the leakage current and shield it from external disturbances. This structure consists of alternating layers of high magnetic permeability and high conductivity materials.
Key methodological aspects include:
- Finite Element Simulation: Used to analyze the relationship between the multi-stage ring's geometry (specifically, the size of the rings) and its magnetic field measurement sensitivity and interference immunity.
- Improved Non-dominated Ranking Genetic Algorithm (ACGWO-BP-NSGA-II): This sophisticated optimization algorithm is employed to find the optimal dimensions of the multi-stage magnetic ring. It combines a chaotic gray wolf optimizer (ACGWO) for nonlinear fitting with the NSGA-II genetic algorithm for multi-objective optimization (balancing measurement accuracy and sensor size).
- Experimental Validation: A custom experimental platform was built to test the performance of the designed sensor under various conditions, including AC/DC measurements and external magnetic interference.
Datasets
Not discussed in the paper. The authors generated simulation data using COMSOL Multiphysics and used experimental data collected from their custom-built platform.
Key Insights and Ideas
- Multi-stage ring design is key: The paper demonstrates that a multi-stage magnetic ring structure, compared to a single-stage one, significantly enhances both the magnetic field amplification and the rejection of external interference.
- Geometric optimization is critical: The sensitivity and anti-interference capabilities are highly dependent on the specific dimensions of the magnetic ring layers, particularly the inner stage polymagnetic layer thickness and the attenuation layer thicknesses.
- Advanced optimization is necessary: The complex, non-linear relationship between the ring's geometry and its performance necessitates advanced optimization techniques like ACGWO-BP-NSGA-II to find optimal solutions that balance multiple objectives (accuracy, size, interference immunity).
- TMR sensors can be made robust: The research shows that with the right structural design, TMR sensors, known for their sensitivity, can be made sufficiently robust for challenging industrial environments.
Key Results
- Improved Sensitivity: The multi-stage TMR sensor exhibits significantly higher magnetic field measurement sensitivity compared to single-stage designs.
- Reduced Nonlinearity: The fitting nonlinearity error is less than 1% for AC/DC measurements.
- Extended Bandwidth: The sensor achieves a measurement bandwidth of 0-80 kHz.
- Lower Detection Limits: The minimum measurable AC current is 85 µA, and the minimum DC current is 50 µA, a substantial improvement over single-stage sensors.
- Enhanced Interference Immunity: The multi-stage design drastically reduces the impact of external magnetic interference, with experimental results showing a significant decrease in measurement error under strong interference conditions. For example, the measurement error (MC) was reduced from 109% for a single-stage ring to 10% for a 2-stage ring under specific interference, and further improved with more stages.
- Optimized Dimensions: The ACGWO-BP-NSGA-II algorithm determined optimal dimensions for a 3-stage ring: r₁ = 10 mm, p = 6 mm, th₁ = 1 mm, th₂ = 8.3 mm, h = 10 mm, and d = 8 mm.
- Error Reduction: The optimized multi-stage structure reduced the measurement error by an average of 105% compared to a single-stage ring of similar volume.
Limitations and Future Work
- Cost and Commercialization: The authors acknowledge that the multi-stage magnetic ring design, without commercialization, does not significantly reduce the sensor's cost.
- Long-term Operation: The hysteresis characteristics of the multi-stage ring structure can impact measurement accuracy during long-term operation.
- Bias Voltage Influence: The TMR sensor's measurement is influenced by bias voltage.
- Future Work:
- Investigate and study the hysteresis characteristics of the multi-stage ring structure.
- Design a relevant demagnetization scheme to further improve accuracy at weak currents.
- Enhance bias voltage elimination circuits and temperature compensation circuits for online monitoring applications.
- Consider adding external shielding shells to adapt to multiple unknown electromagnetic interference sources.
Summary updated