Research On Transformer Winding Deformation Detection Method Based On Distributed Optical Fiber Sensing

Power transformers play an important role in power systems and their safe operation has a direct influence on the reliability and safety of the power supply.Winding deformation is one of the main causes of transformer internal faults.Accurate and real-time deformation detection is of great significance to the safe operation of transformers.Statistics show that insulation damage is the main cause of transformer failure with the winding being the part with the highest failure rate.Contrapose the problem that traditional winding deformation detection method has low sensitivity to local small deformation of windings and cannot locate faults,this paper presents a transformer winding deformation detection method based on distributed optical fiber sensing.The stability of built-in optical fiber in transformer oil is studied.The fitting method of brillouin scattering spectrum is proposed.The strain transfer mechanism between the sensing fiber and winding wire is also studied.Transformer winding deformation detection test based on distributed optical fiber sensing is carried out.The main research contents and results are as follows:Aging test for transformer oil with built-in optical fiber is carried out and the effects of fiber-oil sample and pure oil sample on parameters such as water content,acid value,dielectric loss and volume resistivity are analyzed.The influence of thermal aging on measurement accuracy of optical fiber is analyzed.Polyimide and ethylene-tetrafluoroethylene copolymer(ETFE)were selected as coating and sheath materials of sensing fiber respectively.The influence of two optical fiber installation methods on the electric field distribution inside the windings and the main factors affecting the maximum field intensity inside the transformer windings and optical fiber are analyzed.The power-frequency voltage withstand test is carried out for the fiber composite wires with different installation modes.The results show that the two optical fiber layouts have no effect on the insulation characteristics of transformer.The electrical aging and flashover test of ETFE were carried out.Compared with oil-paper insulation,the electric aging life model of ETFE is studied.The results show that ETFE has better insulation characteristics than insulating paper and meets the insulation requirements of transformer.The wiring and protection method of distributed optical fiber in transformer is proposed and a 35kV oil-immersed transformer with built-in distributed sensing fiber is developed.The Brillouin frequency spectrum fitting method based on cross correlation-LM method and the principle of bimodal fitting are proposed which improving the Brillouin spectrum fitting accuracy and speed.The error is less than 1MHz.Strain and temperature of optical fiber can be measured by using Brillouin-Raman combined method.The accuracy of temperature and strain measurements were±0.5? and±25??respectively.The strain curves and measuring accuracy of the optical fiber composite transformer windings at different internal temperatures are obtained,which verifies the feasibility of distributed optical fiber sensing technology in transformer deformation detection.Two fiber composite transformer winding models are designed and developed.The axial/bending strain transfer model of optical fiber and transformer windings is established.Theoretical research show that the fiber strain transfer rate is more than 85%,which can effectively transfer wire strain.Main factors affecting strain transmissibility are analyzed.The strain transfer function in plastic deformation of wire is proposed.The relation between winding deformation and fiber strain is studied.Compared with the traditional FRA method,the detection sensitivity of distributed optical fiber sensing technology for transformer winding deformation is estimated.The test verities the effectiveness of distributed optical fiber detection method in measuring slight deformation of windings.In addition,a winding deformation detection criterion based on Euclidean distance is proposed.Simulation tests of partial temperature rise and deformation of transformer windings were carried out and realized the winding deformation localization detection.The deformation fault can be located to a certain cake.The simulation tests of transformer windings with different deformation types were carried out and the parameters of fiber optic strain curves are extracted when the windings are concave and bulging in different parts based on S transform.The limit learning machine algorithm is used to identify the fault types of windings.The results show that the recognition of the test samples can reach more than 94%.