4th International Conference on Electric Power Equipment– Switching Technology (ICEPE-2017)

481 / 2017-05-09 20:36:31

Fault Diagnosis for High Voltage Circuit Breaker Based on Hilbert-Huang Transform and Support Vector Machine

High Voltage Circuit Breaker (HVCB),2770,Hilbert-Huang transform (HHT),Support Vector Machine (SVM)

High voltage circuit breaker (HVCB) is one of the crucial equipment in power system, and the power system can be controlled and protected by HVCB. It is necessary to monitor the operation state of the HVCB and analyze the fault by the related signal, the fault of HVCB operating mechanism is closely associated with coil current, mechanical vibration and other factors. Coil current waveform and mechanical vibration signal have different characteristics in different states. A new method of fault diagnosis for HVCB based on Hilbert-Huang transform (HHT) and support vector machine is proposed to analyze faults accurately and quickly in high voltage circuit breakers, determining the characteristic and category of the fault. Compared with Fourier transform-based linear and steady-state spectral analysis, the HHT method can better analyze non-stationary and nonlinear problems. HVCB testing platform based on LabVIEW and NI data acquisition equipment gain the signals of the coil current and mechanical vibration generated by the different fault conditions, these signals are processed by HHT for spectrum analysis, so the normal state and the fault state of the current, mechanical vibration signal and other factors of the characteristics are extracted. The model of fault diagnosis for HVCB is established by using support vector machine (SVM), The characteristics in different states are consider as the input vector of the support vector machine, Data is trained by SVM, and the genetic algorithm is used to optimize the model parameters of the SVM, thus accurate detection of HVCB status is achieved, the type of fault can be clearly determined. This study has carried on the related experiment, the feasibility and the validity of this method has been analyzed and verified. The method can obtain better diagnosis results, it is also suitable for fault diagnosis for HVCB.