Research On Intelligent Detection System Of Radio Signal In Power Cable

With the development of urbanization in China,power cable as an important part of power network is more and more widely used.Although the power cable has the characteristics of high safety and reliability and easy operation,the fault frequency of power cable is increasing with the increasing number of applications and the increasing of operating load and time.Due to the complexity of power cable structure,and they are often laid underground.In case of cable failure,huge economic losses will be caused,if the failure cannot be located and repaired in time.The fast and accurate detection of cable fault location based on the transmission characteristics of electric wave signals in power cables is of great significance to ensure the normal operation of power grid.In view of the problems of large measurement error and low integration degree of existing detection systems,based on the research of signal transmission characteristics in power cables in this thesis,the data processing method combining wavelet analysis and curve fitting is proposed,the pulse generation and signal detection system is designed,and the acquisition,denoising and singularity detection of electric wave signals in cables are realized.The main tasks of the thesis include the following parts.Firstly,the domestic and international status of cable fault location and time-domain reflection measurement technology of pulse reflection method is introduced,and the application of digital signal processing technology in fault location is explained.The transmission characteristics of pulsed electric waves in the cable are analyzed,and the principle of electromagnetic wave time domain reflection is simulated using SPICE software.For the phenomenon of loss of pulsed electric wave signal transmitted in the cable,XLPE medium-voltage single-core cable was taken as the object to study the high-frequency wave propagation characteristics of the cable,and the signal attenuation and dispersion law under high frequency were simulated and analyzed.Secondly,in order to deal with the problem that signals are susceptible to noise and cause large fault ranging errors,a pulse signal threshold noise reduction technique based on wavelet analysis is studied.Combined with MATLAB wavelet analysis tool,sym8 wavelet was used to denoise the measured signal,and the processing results show that it has a good denoising effect.The accuracy of the pulse reflection method depends on the identification of the incident wave and reflected wave front.To reduce the influence of signal attenuation and dispersion,the method of detecting the singularity by the maximum value of wavelet mode is studied.In order to solve the problem of large measurement error of the wavelet modulus maximum method when the signal singularity is weak,a method of combining curve fitting and wavelet modulus maximum value is proposed,and more accurate cable fault location is realized.Finally,a pulse generation and signal detection system for cable fault location is designed and implemented.The system is based on i.MX 6D processor and FPGA.Hardware circuits and FPGA logic such as narrow pulse source circuit,signal conditioning circuit,data acquisition module,ranging mode switching module and SPI communication module are designed.Completed the preparation of Nios II software and Linux device driver,and designed the system GUI interface program and automatic ranging algorithm based on Qt framework.The functions of nanosecond narrow pulse output,signal collection and processing,automatic fault measurement of cable faults,and data waveform display are realized.The software and hardware of the system were tested with a faulty cable,which verified that the system can effectively detect the pulse wave signal and fault distance in the cable.