Feature Optimization And Fault Diagnosisof GIS Based On Combined Detection

Gas Insulated Substation (GIS) is widely used in urban power substation becauseof its advantages of small area coverage, operational reliability and low electromagneticpollution. However, due to some reasons in structure and transportation, there areinevitable insulation defects which may cause partial discharge in regular operation.The safe operation of grid is threatened seriously by insulation damage. Therefore, theon-line monitoring and fault type recognition in GIS has been the focus of research inthis field.In this paper, statistical characteristics of different insulation defects PD signals arestudied based on analyzing researches about PD type recognition home and abroad.Methods to obtain different characteristic parameters are developed from differentangles of partial discharge pattern, Support Vector Data Description method is presentedfor PD type recognition in GIS. The main work and achievements are as follows.①Large data of discharge experiment in different intensity is acquired by PDUHF monitoring system which has been developed for the detection of four kinds oftypical insulation defects model in the laboratory. The φ-u-n3D PD image and thecorresponding φ-u, φ-n two-dimensional image has been constructed.The results showthat: the differences of four PD image shape are rather obvious, the same type of PDimage shape remains unchanged in different voltage levels. Obtaining the13statisticalfeatures based on the two-dimensional image can lay a foundation for the research ofPD type recognition.②Based on the electromagnetic theory, antenna and photoelectron radiationtheory, the relationship between apparent discharge quantity calculated by IEC60270and the detection signal could be obtained. Linear relationship between energy of UHFsignal and square of apparent discharge quantity has been found, while linearrelationship between the integral value of the signal of the optical method and thecharge quantity comes out. According to the practical situation, the correction formulaof gas pressure has been get by the research on the influence of different gas pressure toUHF signal and apparent discharge quantity. The effectiveness of optical method hasbeen also proved by experiment result.③The KPCA method is presented for the extraction of feature subset of fourkinds of insulation defects discharge comprehensive characteristic. The Maximal Relevance Minimal Redundancy method is put forward for dimension reduction of theoriginal statistical characteristics. With apparent discharge quantity q, energy of UHFsignal E and optical energy A, two algorithms are combined to construct optimal featuresubset TBESTto identify four kinds of defects, which effectively retained thecharacteristics of each class of the PD signal and reduced the redundancy of statisticalcharacteristics.④SVDD is introduced into PD type recognition, based on the principle ofMaximum interval of support vector machine and one to multiple of multipleclassification method, an optimal radius support vector data description algorithm(OR-SVDD) is proposed to solve the disadvantages of missing and wrong classificationin SVM and the lack of classification margin small in SVDD. The principle of“one-to-many” is adopted to solve the difficult problem of multi-class defectclassification, and to improve the identification ability and application value.⑤As a result of optimization of classification performance, local GA algorithmand SA algorithm have been applied to optimize the kernel parameter σ and the penaltyfactor C of SVDD classifier. According to the real project, PCA-FDA has been proposed.Two overlapping classes of fault would desperate in original space, and then reach themaximum level in Fisher projection space. With the method, the low recognition ratio60%would rise high to80%to improve the applying value.