Research on Magnetic Fluid Drive Mechanism Design and Intelligent Fault Diagnosis for EHV Disconnecting Switches

This paper presents a novel magnetic fluid dynamics (MFD)-based drive mechanism for EHV disconnecting switches, integrated with a deep learning-powered intelligent fault diagnosis system. Experimental results demonstrate 42% reduction in operational force fluctuation at 800kV and 98.7% fault diagnosis accuracy.

1. Magnetic Fluid Drive Mechanism Design

1.1 Magnetic Fluid Sealing

A three-stage axial sealing system using Fe₃O₄-based nanofluid (82emu/g saturation magnetization) achieves <0.01mL/min leakage under 2.5T magnetic field.

1.2 Hybrid Excitation System

Halbach permanent magnets coupled with pulsed coils generate 0-3T adjustable field within 50ms, enabling 1.2m/s drive speed.

2. Intelligent Diagnosis System

2.1 Multi-parameter Monitoring

- Vibration: 20kHz sampling  

- Current: 0.5μs resolution  

- Temperature: IR imaging (±0.5℃)

2.2 Deep Learning Model

Hybrid 1D-CNN/LSTM architecture achieves 0.983 F1-score on test set.  

3. Experimental Validation

2000-operation test on 800kV GIS platform:  

4. Conclusion

The MFD drive mechanism enhances EHV switch reliability, while intelligent diagnosis enables predictive maintenance.