基于嵌入式机器视觉识别的电力设备监测系统设计

Operation Monitoring System Design of Power Equipment Based on Embedded Machine Vision Recognition

为了提高对电力设备运行过程的监控能力,设计了嵌入式的机器视觉系统,该系统包括图形采集模块、图像处理模块、图形显示模块、结果输出模块等。通过嵌入式技术设计,实现了电力设备图像信息的采集、处理、显示和输出;通过OV7670图像传感器提高了图像采集能力。在进行故障信息诊断时,采用基于多主成分分析模型及支持向量机-DS融合决策的方法,实现正常数据信息与故障数据信息的分离。实验表明,该系统故障诊断正确率高达95%,大大提高了对电力设备监控能力和故障处理效率。

In order to improve the ability of monitoring the running process of power equipment,an embedded machine vision system is designed,it includes graphics acquisition module,image processing module,graphics display module and result output module.Through the design of embedded technology,the acquisition,processing,display and output of power equipment image information are realized.Through the OV7670 image sensor,the image acquisition ability is improved.When diagnosing fault information,based on multi principal component analysis model and support vector machine-DS fusion decision method,the separation of normal data information and fault data information is realized.The test results show that the fault diagnosis accuracy of the system is as high as 95%,which greatly improves the monitoring ability of power equipment and fault processing efficiency.