Research Of Distribution Network Fault Indicator Based On Compressed Sensing

Distribution network is user-oriented terminal in the supply chain of the power grid.People's growing demand for electricity requires a more stable and sufficient electricity supply.The developing of renewable energy contributed to more complex distribution network structure,and the huge amount of sampling data of the distribution network both bring difficulties to the rapid fault location.Fault indicator based on the compressed sensing,can reduce about a half of sampling points of the same frequency signal,which reduce the amount of data greatly and accelerate the speed of data processing.On the basis of the research with compressed sensing sampling method,this paper proposes a non-configuration recovery algorithm with integrate and fire sampling method,which can accelerate a half of sampling time.At the same time,the simulation model and hardware design are introduced.The main contents are as follows:Firstly,the compressed sensing theory is introduced.The advantages of compressed sensing sampling in data volume collection are concluded by building the simulation model and sampling with the common electric signals.After comparing two compressed sensing sampling method,this paper selects the integrate and fire sampling method as the sampling method of the fault indicator.Then,sampling with single-frequency signal and multi-frequency signal,the relationship of the output pulse sequence?model paraments?waveform frequency and amplitude is analyzed.Besides,a non-configuration recovery algorithm with integrate and fire sampling method is designed.The model of electric transmission line with single-phase earth fault is created in the Simulink and used for analyzing the method of fault occurred line and phase detecting on the basis of non-configuration recovery algorithm with integrate and fire sampling method,which proved the validity of fault indicator with compressed sensing sampling method.Finally,in consideration of the actual situation of the line,the module required for the actual operation of the fault indicator is designed and test.The Proteus software is used for simulation design of the CT power module and the current-voltage conversion module.In addition,the CT power module adds a voltage stabilizing module to prevent the chip from being damaged by impulse current and fault current.The current-voltage conversion circuit is built with an operational amplifier,and the appropriate conversion coefficient is obtained under the condition of small input resistance.The Integrate and fire sampling module includes modules such as rectifier circuit,integrator,and comparator.It is proved that output pulse fluctuate with time because the noise interference,but distribution and quantity of output pulse are same.The fault judging program is designed with STM32-F407 as the core,and LED can be lighted when inputting the fault waveform,by sampling the amplitude and direction features and judging by the fault criterion.