Research Of Arc Fault Detection For AC Solid State Power Controllers Based On Wavelet Transform

Solid state power controller(SSPC)is the key component of advanced aircraft power distribution system and is adopted to control and protect the transmission lines intelligently.In addtiton to the common faults such as overcurrent and short circuit,arc fault will also damage the transmission lines in aviation electrical system.In order to improve the overall safety of the aircraft,it is necessary to study the SSPC with function of arc fault detection(AFD).In this paper,arc’s discharge characteristics and the generating and classification of fault arc are introduced firstly,and the volt-ampere characteristics,time-domain characteristic and frequency-domain characteristic are summarized.The simulation of serial AC arc with different types of loads is completed based on Mayer arc model,which can be used to provide the theoretical basis for the feature extraction and criterion of AC arc fault.Secondly,the research status of AC AFD method is summarized.A terminal loosing experiment is conducted in the laboratory to obtain the arc fault waveforms of pure resistor load and resistor-inductance load.Based on the wavelet modulus maxima and singularity detection principle,three-layer Db3 wavelet transform is used to extract the eigenvalues of normal current and arc current.The result shows that each layer’s detailed waveform coefficient of arc fault signal is greater than that of the normal signal.After analyzing the fast achievement method for wavelet transform and studying the feasibility of SWT hardware implementation,arc fault detection criterion of AC SSPC is established as that the detection unit is half-period AC wave and the eigenvalues are the ratio of falt shouler and singular value according to the characteristics of AC arc fault in time domain and time-frequency domain.Hardware and software design of every functional modules are completed in compliance with the requirement of AC SSPC arc fault detection.Hardware design includes power supply circuit,signal conditioning circuit and frequency detection circuit.Software design includes the development process and description of subprograms.Finally,experimental verifications are carried out on crucial hardware modules and program segments based on Infineon XE164 FN microcontroller.The experiments of conditioning circuit,frequency detection circuit and triangular wave sampling using the analog to digital converter are accomplished.The correctness and real-time of SWT program are verified and the least number of half-wave in this system is confirmed,which lay a foundation for the following work.