基于计算机视觉的管道保温破损识别方法研究  

Research on identification method of pipeline insulation damagebased on computer vision

作　　者：高丽岩 Gao Liyan(Equipment Research Institute of SINOPEC Tianjin Company,Tianjin 300270)

机构地区：[1]中石化(天津)石油化工有限公司装备研究院,天津市300270

出　　处：《炼油技术与工程》2024年第7期50-52,共3页Petroleum Refinery Engineering

基　　金：中国石化课题“管道保温分级管理技术开发”(CLY18029)。

摘　　要：针对炼化企业中管道保温层散热损失问题,提出一种利用计算机视觉识别炼化企业管道保温破损并定位的方法。通过无人机搭载红外热成像技术进行图像采集,利用ResNet50架构进行特征提取,并结合Faster RCNN神经网络实现破损的准确识别与定位,阐述了图像预处理、数据集构建、模型训练及预知性维修等关键步骤,通过深度融合无人机与计算机视觉技术,显著提升了检测效率和精度。此外,利用生成的保温破损分布定位图,企业可制定数据驱动的维护策略,降低运维成本并延长管道保温使用寿命,从而提高整体运营效率,为管道系统的长期安全稳定运行提供了科学支持。A method using computer vision to identify and locate pipeline insulation damage in refining and chemical enterprises is proposed to address the issue of heat dissipation loss in pipeline insulation layers.The key steps of image preprocessing,dataset construction,model training,and predictive maintenance are explained by using infrared thermal imaging technology on drones,feature extraction using ResNet50 architecture,and Faster RCNN neural network to accurately identify and locate damages.Through deep fusion of drones and computer vision technology,detection efficiency and accuracy are significantly improved.In addition,using the generated insulation damage distribution and positioning map,enterprises can formulate data-driven maintenance strategies,reduce operation and maintenance costs,and extend the service life of pipeline insulation,thereby improving overall operational efficiency.This provides scientific support for the long-term safe and stable operation of the pipeline system.

关 键 词：管道 保温破损 红外热成像 图像特征 目标识别 神经网络 定位检测 预知性维修 

分 类 号：TE682[石油与天然气工程—油气加工工程] TP183[自动化与计算机技术—控制理论与控制工程] TP391.41[自动化与计算机技术—控制科学与工程]