Detection of Faulty Vibration in Oil-Immersed OLTC by Fiber Bragg Grating Based on Double Cantilever Beam Type Sensitization

Fiber Bragg grating sensing offers advantages, such as resistance to electromagnetic interference, small size, strong multiplexing and wide detection bandwidth, which can be used to detect on-load tap changer (OLTC) fault vibration detection. This paper designs a fiber Bragg grating (FBG) vibration sensor based on a double cantilever beam amplification structure. The designed sensor adopts a two-point adhesive packaging structure, and theoretical analysis shows that the resonant frequency of the Bragg grating vibration sensor is 3200 Hz. The goodness of fit ( $R^{2}$ ) of the two designed sensors are 0.9993 and 0.9996, respectively, and the sensors can effectively detect vibration when the acceleration varies from 2 m/s2 to 20 m/s2. A real oil-immersed OLTC fault simulation platform was built at Shandong Taikai Company, and vibration detection was carried out for four types of faults. A comparison of internal and external fiber Bragg grating sensors for detecting OLTC vibration signals revealed that the signal-to-noise ratio of the fiber Bragg grating sensor when embedded in the inner cylinder was significantly higher than when placed in the outer cylinder, and the signal amplitude was also higher in the inner cylinder. This is because during the propagation of the vibration signal, the signal traveling from the inner cylinder to the outer cylinder leads to significant attenuation of higher-frequency vibration signals, causing signal distortion. When the drive shaft bolt is loose or the arc-shaped plate is loose, the signal frequency mainly ranges from 500 Hz to 750 Hz, and the energy in this frequency band exceeds 40%. When contact wear or spring fatigue occurs, the energy mainly concentrates in the 0 Hz to 250 Hz range, with energy exceeding 35% in this frequency band. The spectrum allows different fault signals to be distinguished, laying a solid foundation for subsequent fault mode recognition and fault severity classification of OLTC.