Analysis And Phase Identification Of Air Gap Discharge Characteristics Of Superconducting Cable Composite Insulation In Liquid Nitrogen Temperature Region

High temperature superconducting power transmission has the advantages of large capacity,low loss,small volume and high system reliability,which is expected to play an important role in the future development of China’s power grid.The cryogenic insulation technology is one of the key technologies in the development of superconducting power transmission.Superconducting cable insulation materials manufacturing process generally around the package method is used to reduce the dielectric loss,the selection of winding and tension and prone to micro cracks and pores,the processing of detail in all kinds of stress under the joint action of accelerated crack easily,in a large electric field intensity under the action of partial discharge is given priority to with air gap discharge,composite insulation may face the risk of failure.Although the high-temperature superconducting cable transmission technology has not yet reached the stage of large-scale application,it is an important link to promote the application of superconducting cable engineering to solve the insulation problem under operating conditions.Therefore,the research and analysis on the development characteristics of composite insulation air gap discharge of superconducting cables is helpful to improve the reliability of on-line monitoring of composite insulation of superconducting cables,and is of great significance to the safe operation of the whole power grid.In this paper,an experimental platform for composite insulation gap discharge was built based on the operating environment of the simulated superconducting cable in the laboratory.A large number of gap discharge signals under different voltage levels were obtained by using the pulse current method detection system.The development characteristics of composite insulation gap discharge under normal temperature and liquid nitrogen temperature were analyzed.The statistical operator and the variational mode decomposition method were used to extract the characteristic parameters of air gap discharge,and the dimensionality reduction was carried out to obtain the 8-dimensional characteristic parameters.Finally,an improved KNN method is proposed to identify the development stages of composite insulation discharge by using hierarchical clustering method.The main achievements are as follows:(1)An air gap discharge feature extraction method based on variational mode decomposition(VMD)is proposed.The air gap discharge signals of composite insulation under liquid nitrogen temperature region were decomposed by variational modes.The shape and texture features of the decomposed k modal component images were obtained by using the relative moment and gray level co-occurrence matrix,and the multi-scale entropy values were extracted to form the discharge feature vectors.(2)An improved KNN method is proposed to identify the air gap discharge stage of composite insulation.Will composite insulating air gap discharge four stage,a total of four categories as output,the composite insulation stage of development for effective identification,and improved KNN and kernel function based on binary tree support vector machine(SVM)is compared,the results show that after weighted KNN recognition rate average recognition rate is highest,and the effectiveness of this method has a certain.