Research On The Method And Device Of DC Arc Fault Detection

As the DC power system is widely used in many fields such as aerospace,ships,power grid systems,etc,with the rapid development of Power electronic technology.The DC arc fault is easily caused by the aging wires,loose connections and other reasons.Since current through a series DC arc does not possess a periodic zero crossing,it is much more likely that an arc in DC system will result in more sustained ignition compared an AC system and easy to cause fire hazard.Therefore,the series DC arc fault is a serious threat to the security and stability of the DC system.Through the analysis of the frequency characteristics of dc arc current,Dc series arc fault detection method and the hardware device are studied by us.According to the standard of UL1699 b,series dc arc fault test platform is build and the arc voltage and current data is collected.A recognition method of series arc fault is put forward based on ensemble empirical mode decomposition(EEMD)according to the characteristics of the arc current in frequency domain.Firstly,the arc current signals are processed by EEMD to acquire a set of intrinsic mode function(IMF).Secondly,IMF5 and IMF6 are selected to represent appearance of a series arc fault through analysis of the relationship between the IMFs and arc fault.Finally,the setting threshold values for IMFs are calculated and decided.It is concluded that IMF5 and IMF6 increase significantly during the arc fault and it is detected accurately accordance with the setting threshold values,which means the presented EEMD algorithm is feasible for DC arcing fault identification.A sliding DFT based arcing fault recognition method is proposed to meet the requirement of real-time in detection and reduction in algorithm complexity.The arcing initiation stage is the best moment for fault detection based on analysis both in time domain and frequency domain.The characteristic frequencies,including 3 KHz,5 KHz,40 KHz,80 KHz and 100 KHz,are selected for the point-by point sliding DFT calculation.And a time-window of 200?s is used to implement the algorithm.And hardware device is detected by using field programmable gate array(FPGA).It is concluded that all the spectrum of characteristic frequencies increase significantly during the arcing fault.And it is verified that the sliding DFT algorithm is feasible for DC arcing fault recognition.