一种ZYNQ的GIS设备局部放电检测系统设计

Partial discharge pattern recognition of GIS equipment based on ZYNQ

为有效检测局部放电信号并识别放电类型,利用多核异构处理器ZYNQ PL侧高速并行处理能力,设计了一种使用ADC采集局部放电信号,并通过深度学习识别放电类型的局部放电检测系统。介绍了系统的硬件设计以及逻辑设计。系统PL侧通过外部工频同步电路触发对局部放电信号的采集,并将采集数据传入PS侧进行数据分析处理,最后使用深度学习对放电类型进行分类。实验结果表明,该系统能够有效采集局部放电信号并识别放电类型。

In order to effectively detect partial discharge signals and identify discharge types,using the high-speed parallel processing capabilities of the multi-core heterogeneous processor ZYNQ PL side,a partial discharge detection system is designed that uses ADC to collect partial discharge signals and identifies discharge types through deep learning.The article introduces the hardware design and logic design of the system.The PL side of the system triggers the collection of partial discharge signals through an external power frequency synchronization circuit,and transmits the collected data to the PS side for data analysis and processing.Finally,deep learning is used to classify discharge types.Experimental results show that the system can effectively collect partial discharge signals and identify discharge types.