Research On Physical Mechanism Of The Uniformly-distributed Equal Ampere-turns Verification Method Of Current Transformers

In three phase system, the serious influence of external currents on the current transformer's accuracy shouldn't be ignored. Domestic authorities presented the uniformly-distributed equal ampere-turns verification method, which concentrates the primary turns on one part of the current transformer's core, to measure the error of current transformer in actual working environment. However, there's few or no special research on this verification method at home and abroad up to now. Therefore, the research on the theory of this method will be valuable and significant for verification, design and improvement of current transformer.Firstly, this paper uses equivalent circuit to analyze the influence of external current's stray magnetic field on the error of current transformer. Then, this stray magnetic field is solved with separation of variables and finite element method separately. Compared with actual measured value, the numerical computation result of three-dimensional magnetic field proved accepted. It indicates that the influence quantity of external currents on current transformer depends on the magnitude of stray flux. If the resultant flux of stray and working flux of current transformer is big enough to make the current transformer's core saturated, the accuracy of current transformer will consequentially get bad. Otherwise, the change of current transformer's accuracy could be ignored.Secondly, this paper takes an example to analyze the influence of concentrated primary turns'leakage magnetic field on the error of current transformer. Also, this leakage magnetic field is solved with finite element method. Like the stray flux, the influence quantity of leakage magnetic field on current transformers depends on whether the current transformer's core is saturated in the resultant magnetic field of leakage and working magnetic field. The change of primary turns'concentrated angle on current transformer's core could change leakage magnetic field.The radial basis function neural network is used to fit the numerical computation results of stray flux and leakage flux separately. Then these two radial basis function neural networks are combined to give concentrated angle of primary turns on the current transformer's core to ensure that flux maximum of the core under verification equal to flux maximum in actual working environment. The measurement result is the error of current transformer in actual working environment consequently. An example is taken to prove that computational method is rational.