| Cable tunnel power equipment online monitoring systems often need to collect multiple power equipment signals.Since each monitoring subsystem is independent of each other,multiple data cannot be processed in a unified manner.Multiple processing units need to be deployed,which not only increases the cost,but also leads to the cable monitoring system The increased complexity of the monitoring system will also affect the stable operation of the monitoring system.In order to solve the above problems,this paper designs a set of integrated live detection equipment and auxiliary system for power pipe gallery based on partial discharge signal analysis and intelligent detection to realize cable online partial discharge detection.This paper analyzes the law of partial discharge of power cables,signal analysis of partial discharges,and detection of partial discharges to realize the accurate positioning of cable partial discharges,and develops cable live detection devices based on FPGA.The work of the thesis mainly includes:1.Study the discharge law of cable partial discharge,focusing on the characteristics of ethylene-propylene rubber cables and various parameters that affect the partial discharge signal;then analyze the partial discharge PRPD spectrum,establish the cable partial discharge discharge model,and analyze its in the power cable The attenuation characteristics of the cable provide a theoretical basis for the online detection of cable partial discharge below.2.Considering the error and noise problems of cable partial discharge signal collection,it must be dealt with.First,the integrity of the PCB signal is optimized,and the PCB signal collection is realized from three aspects: signal reflection,signal delay and signal crosstalk.Process,and then perform wavelet denoising and intensity judgment on the collected signals to ensure the availability and accuracy of the collected partial discharge signals.3.Based on the deep convolutional neural network algorithm to realize the cable partial discharge detection and positioning,according to the influence of different influencing factors on the convolutional neural network partial discharge identification,analyze the ability of the convolutional neural network to perform partial discharge detection under different parameters,focusing on the analysis before and after To update the parameters of the propagation,and analyze the differences in the number of convolutional layers,activation functions,and optimization algorithms to achieve partial discharge detection,indicating the effectiveness and efficiency of the algorithm in this paper.4.Based on SoC FPGA for cable live detection device development,the overall function design of the device is carried out on the ZYNQ series platform,and the high-frequency pulse current sensor(HFCT),power frequency phase transformer,ground current transformer,and high-speed acquisition card The parameters of the template such as the motherboard and the motherboard are designed,and the high-speed AD chip is used to achieve high-speed collection of cable signals,and then the previous algorithm is used to identify and process the collected signals to realize the partial discharge live detection of the cable. |
