摘要
针对运检人员难以及时作出判断并进行维修的问题,将增强现实(Augmented Reality,AR)和深度学习相结合,设计了一种基于AR-OP-CNN-GRU的电力设备故障检测模型。基于AR技术从数据层、服务层和应用层构建出电力设备故障整体诊断模型,再利用自适应中值滤波方法对图像进行去噪,将最大类间方差法(OTSU)和脉冲耦合神经网络(Pulse Coupled Neural Network,PCNN)结合快速提取故障区域。同时,提出了一种电力设备故障识别算法,构建出融合了局部特征预提取模块的CNN-GRU故障诊断模型。算例分析结果表明,所提AR-OP-CNN-GRU电力设备故障检测模型能够对虚拟电力设备的故障进行快速精确的检测识别。
A power equipment fault detection model based on AR-OP-CNN-GRU is designed by combining Augmented Reality(AR)and deep learning to address the issue of difficulty in timely judgment and maintenance by inspection personnel.Based on AR technology,a comprehensive diagnosis model for power equipment faults is constructed from the data layer,service layer,and application layer.Then,adaptive median filtering method is used to denoise the image.The OTSU method and Pulse Coupled Neural Network(PCNN)are combined to quickly extract the fault area.At the same time,a fault recognition algorithm for power equipment is proposed,and a CNN-GRU fault diagnosis model integrating local feature pre-extraction modules is constructed.The numerical analysis results show that the proposed AR-OP-CNN-GRU power equipment fault detection model can quickly and accurately detect and identify faults in virtual power equipment.
作者
李建路
朱珠
李柯
王振乾
盘晴
LI Jian-lu;ZHU Zhu;LI Ke;WANG Zhen-qian;PAN Qing(China Southern Power Grid Energy Storage Co.,Ltd.,Guangzhou 510630,China)
出处
《信息技术》
2024年第12期136-140,148,共6页
Information Technology
基金
南方电网调峰调频发电有限公司信息通信分公司科技项目(022100KK52190003)。
关键词
智能感知
增强现实
检测识别
卷积网络
门控循环单元
intelligent perception
Augmented Reality
detection and identification
convolutional network
gate control loop uni