| The high-voltage switchgear is the main equipment for controlling the on-off state of the line,and its state monitoring is of great significance to improving the stability of the power system.With the development of detection technology and the advancement of the Internet of Things,local processing of discharge signals has become a new research trend.In view of this trend,this paper conducts a lot of research and selection on partial discharge detection technology and data processing algorithm.At the same time,this paper selects FPGA(Field Programmable Gate Array)as the core algorithm to realize the hardware platform,which has powerful data processing capabilities,flexible.Based on the defect characteristics and actual operation of the switchgear,this paper firstly extracts three feasible main detection methods: UHF(ultra high frequency,UHF)detection method,ultrasonic detection method,and transient grounding detection method,and designs three different defect types.Considering the advantages and disadvantages of various detection methods,in recent years,more and more scholars have gradually adopted the joint detection method to detect and monitor electrical equipment.Among them,UHF detection method and ultrasonic detection method are commonly used in high-voltage switchgear.detection and further identification.Afterwards,the PD signal types and characteristics of the detection technology are analyzed,and the algorithm is analyzed.The UHF detection method detects the electromagnetic wave radiated by partial discharge with high frequency.In order to reduce the difficulty of signal acquisition,a detection circuit is designed to extract important information such as amplitude,mean and phase,and realize the PRPD(PRPD Phase Resolved Partial Discharge)map is characterized by terminal processing and transmitted to the host computer for display.Ultrasonic detection method detects the ultrasonic frequency band of partial discharge,which is convenient for sampling chip acquisition.At the same time,with the improvement of ultrasonic sensor sensitivity,on the one hand,ultrasonic signals containing a large amount of information can be detected,and on the other hand,it is possible for the hardware platform to analyze a single signal.In the hardware,the wavelet packet transform is used to process the signal and extract the characteristic quantity in the time-frequency domain.In order to further study the realization of data processing on digital chip terminals,this paper designs a hardware platform with FPGA as the core,with partial discharge signal conditioning circuit,analog-to-digital conversion circuit,data transmission and other modules.processing,transmission,etc.In the software design,a map feature extraction module and various control modules are built.After the UHF detection signal is processed,the extracted data is uploaded through NB＿IOT.The displayed map verification shows that the feature algorithm implemented by hardware has high accuracy and effectiveness.Identify the type of discharge.At the same time,a feature extraction module based on wavelet packet is designed to extract the energy fractal characteristic parameters of ultrasonic signals.The test shows that the function of the module is correct and the data error is small.And the data of the switchgear detection terminal has great reference significance for the design of geochemical processing. |
