Research On Signal Processing Method Of The Online Calibration System For Current Transformers

As an important measuring element in power system,current transformers are the data sources of various measuring instruments and relay protection equipment.In order to ensure the accurate and stable measurement of current transformers,it is necessary to calibrate them regularly.On-line calibration of current transformer can truly reflect its actual operation and avoid power outage.To ensure the continuous power supply in the calibration process,it is necessary to use the opening type current transformers.However,due to the opening air gap and the limitation of practical conditions,the output accuracy of opening type current transformers is difficult to meet the requirement of calibrationIn this paper,the principle of opening type current transformer was introduced firstly.The noise composition of the output signal was analyzed by Allan variance method.And the digital integration algorithm was selected.On this basis,combined with the frequency fluctuation and noises of power system,an extended Kalman filter method based on colored noise was proposed,and the effectiveness of the method was verified by a simulation example.Fixed noise variance was adopted in this method.But in practical application,the noise variance may be not fixed.To solve this problem,an opening current transformer signal processing method was proposed,which used neural network to modify the output of Kalman filter.The simulation results in random noise environment showed that this method can still maintain good filtering performance and have strong adaptability when the statistical characteristics of noise change.According to the above theory,the signal processing program of opening type current transformer was compiled in LabVIEW and MATLAB,and the experimental platform is composed of standard current transformer,opening type current transformer,data acquisition card and AM-FM voltage source.The experimental results showed that the method can effectively improve the accuracy of the output signal of openinging type current transformer.