Research On Distribute Detection System Of Switchgear Partial Discharge

With the development of the society, the requirement for the quality of powersupply is more and more higher. So high-voltage switchgear has more and moreapplications in the power system due to its characteristics advantage, such as smallvolume, high quality, etc. Operation experiences show that intrinsic defects inswitchgear still cause accidents though its high reliability. Insulation breakdown isprimary reason for switchgear electrical failures. The cause of failure in high-voltageswitchgear was mainly in insulation, conductive and mechanical aspects. By PD on-linemonitoring systems, high-voltage switchgear internal defects can be discovered in timeto avoid the occurrence of fatal failure.At home and abroad partial discharge detection has done a lot of work, sensortechnology and anti-jamming technology also need research. On the basis of analysis ofthe research status of PD online monitoring, a UHF antenna sensor was designed andfusing morphology-complex wavelet transform denoising method was discussed, then,we got the PD detection system design programs that was based on the Androidplatform. The main content of this paper are the following:①It is difficult to suppress the periodic narrowband interference and white noisein the Partial Discharge (PD) monitoring. A suppression method of interference signalswas proposed in this paper. It was based on the fusing morphological filters andcomplex wavelet transform. Based on the principle of mathematical morphology, thegeneralized morphological filters was structured and set as the pre-filter unit to realizethe pretreatment of the original PD signals. The complex wavelet transform was used toprocess PD signals outputted from the pre-filter unit, and the PD signals were obtained.The proposed method was applied to denoise the noise in the simulation PD signals andthe acquisition PD signals. The results show that the applied method can restrain theperiodic narrowband interference and the white noise in PD signals, effectively.Compared with the wavelet denoising method and the complex wavelet denoisingmethod, the applied method can decrease energy loss and retain the PD signalscharacteristics better.②This paper designed the whole framework of Android PD detection systembriefly and completed the hardware system design. The Ultra High Frequency (UHF) detection module was designed for that the requirements of UHF method in highvoltage switchgear PD detection. It mainly included the antenna sensor,filter-amplification unit, high-speed acquisition unit, the wireless WiFi unit and powerunit. In the detection module, the antenna sensor and filter-amplification unit wasstudied mainly. According to the simulation and laboratory test results, we can find thatthe sensor and filter-amplifier meet the requirements. The software part of the systemwas developed, including data acquisition, monitoring program, user management,history query module, parameter setting, etc.③The PD detection system was tested in lab, and four kinds of typical PD defectmodel was made to simulate switchgear in the possibility of corona discharge, internaldischarge, suspended discharge and creeping discharge. To verify that the UHF sensordetecting effect we made current sensor as a reference to collect PD signal, set theinterphone, FM wireless transmitter, fluorescent lamp and the old air conditioning fourkinds of interference environment. Final the test result was good, and, PD detectionsystem can be stable operation.Through the above analysis and test results, the PD detection system based onAndroid platform was research and development in this paper, that had a good effect tohigh voltage switchgear partial discharge detection, software upgraded and equipmentcarrying convenience. There was a good prospect of application.