Abstract:Oil-immersed current transformers (CTs) are key components in power systems, and accurate detecting of their internal temperature and pressure is essential for ensuring safe grid operation. To overcome the limitations of traditional electrical sensors, such as susceptibility to electromagnetic interference and difficulty in multi-parameter detection, a temperature-pressure composite sensing scheme for oil-immersed current transformers based on fiber Bragg grating (FBG) is proposed. First, the relationship between the FBG central wavelength and variations in temperature and pressure is theoretically derived, clarifying the influence mechanism of these parameters on wavelength drift. Then, considering the structural characteristics of the oil sampling valve of a 110 kV oil-immersed current transformer, a composite sensor integrated with the oil sampling device is designed. A dual-FBG configuration is employed to achieve decoupled detection of temperature and pressure, in which FBG1 combined with a circular diaphragm structure senses pressure, while FBG2 encapsulated in the central through-hole of the packaging rod is only sensitive to temperature, and the temperature compensation grating method is used to eliminate temperature cross-sensitivity. Finite element simulation results show that a beryllium bronze diaphragm with a radius of 9 mm and a thickness of 0.3 mm can satisfy the detection requirements under both normal and fault operating conditions. Finally, a prototype sensor is fabricated and an experimental platform is established to verify its performance. Experimental results indicate that the linear fitting coefficient between wavelength drift and temperature exceeds 0.995, with an average temperature sensitivity of 34.96 pm/℃, while the average pressure sensitivity reaches 6.475 pm/kPa with a maximum error of 3.819 kPa. The proposed sensor can be installed without modifying the original structure of the current transformer, overcoming the limitation of traditional single-parameter detection and providing a reliable approach for the condition detecting of oil-immersed current transformers.