Research On The Key Problems With Reliability And Long-term Stability Of Air-core Coil Current Transformers

As an important equipment in smart substations,the performance of air-core coil current transformers directly relates to the reliability of protective relays and the accuracy of the measuring,metering device.However,compared with traditional current transformers,the fault rate of air-core coil current transformers is obviously higher.The failure phenomenon includes the output reliability of air-coil current transformers during disconnecting switch operation and the long-term stability of the error state under the action of environmental parameters.The aim of this paper is to reveal the essence of the problem with the reliability and long-term stability of air-core coil current transformers,discover and tackle the problem of air-core coil current transformers.To this end,the functional mechanisms of the influencing quantities such as disconnecting switch and environmental parameters are investigated intensively,and the analyzing methods for error state of transformers are established.The key research points and achievements of this paper mainly include:（1）To ensure the reliability of air-core current transformers under complicated electromagnetic environments,the transient characteristics of transformers during fast transient is modeled and simulated,and the anti-interference measures of transformers are proposed.To solve the issue of the applicability of lumped-parameter modeling methods under high-frequency operating conditions,a modeling method is put forward for the transient characteristics of transformer based on distributed-parameter model.Then the features of the output signal of transformer can be obtained based on the distributed-parameter model,GIS model and disconnector switching model.The effectiveness of this modeling method is verified on a 220 kV disconnector operation experiment.Besides,to bypass the problem with electromagnetic protections against strong transient multi-route coupling interferences,an influencing model of the disconnector switching is developed,and some anti-interference measures such as charge-discharging circuit,decoupling of the common impedance coupling,cascading filtering circuit,and ground wire separation are presented.These measures are validated on an 1000 kV disconnector operation experimental platform.（2）The error characteristics and influencing factors are systematically studied to manifest the connotation of problem with the stability of error state of transformers and provide guidance for the design and manufacturing of transformers.According to the restrictive relationship between the coil former and the winding of transformers,the comprehensive impact of temperature on transformers is analyzed using the independent error synthesis method.To develop the influencing model of vibration and magnetic field,the discontinuities of the non-uniform air-core coil is equivalent to a region without windings,this method could overcome the constraint of the radius of coil windings on the discontinuities.With regard to the problem of the influencing factors of off-center error,a sensitivity analysis method is introduced independent of the mechanism models of air-core coil current transformers,where the importance of different influencing factors could be identified according to the variance analysis results of orthogonal test.The effectiveness of this analysis method is demonstrated with an off-center error experiment.The results show that the impact of off-center distance is the most significant,the impact of relative position is minor,and the impact of discontinuity is the minimum.（3）To control and evaluate the stability of the error state of air-core coil current transformers in operation,the correlation of the transformers error with the environmental parameters is discussed.The parameters involve as temperature,humidity,vibration,magnetic field and load current.To handle the problem with the decoupling of the error and multi-coupling environmental parameters,a transformers error state correlation analysis method is proposed based on the random-matrix theory.The inner relation between the error and the environmental parameters is quantified as two correlation evaluation indexes,i.e.d_MSRSR and I_MSR.The correlation degree can be acquired in real-time in view of the value and changing trends of the evaluation indexes.The operational data of a transformer state monitoring platform in a 220 kV substation are analyzed using this method,the results indicate that the ratio error of the transformer has strong correlations with temperature,magnetic field and load current,and weak correlations with vibration and humidity.On the other hand,the phase error of the transformer has strong correlations with electrical parameters,and weak correlations with non-electrical parameters.（4）Predicting the changing trends of transformer error is advantageous for the maintenance personnel to track the long-term stability problem of transformer error state.Aiming at this problem,a predicting model for the error of air-core coil current transformers is constructed.To avoid the influence of the abnormal sample data on the structure and parameters of the predicting model,the distribution of transformer error is tested using K-S method,then the abnormal value of the sample data with non-normal distribution is eliminated based on the data interval withβ（g,h）distribution.Since there is not a definite functional relationship between the transformer error and the environmental parameters,a parametric model for predicting transformer error using multidimensional influencing quantities regression algorithm.To upgrade the generalization ability of the predicting model,the sample data is processed with clustering analysis,then the central vectors and weighs of RBF neural network is determined by training the processed data to develop a non-parametric model.The error of an air-core coil current transformer is predicted using the developed predicting model,the results show that these two models could be used to predict the error of transformers in operation,wherein the phase displacement prediction error of parametric model is minor and the ratio prediction error of non-parametric model is relatively smaller.The research achievements of this paper could provide theory and technical supports for the anti-interference design of air-core current transformers,and the protection malfunction phenomenon due to the reliability of transformers can be reduced.Meanwhile,the methods presented could offer effective solutions to analyze the operational trends and implement active maintenance and management,as a result the risk of power energy trading can be prevented.