基于ResNet改进的管道缺陷量化方法研究  

Research on Pipeline Defect Quantification Method Based on ResNet

作　　者：吴国栋 何璐瑶[1] 杨理践[1] 刘斌[1] WU Guodong;HE Luyao;YANG Lijian;LIU Bin(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China)

机构地区：[1]沈阳工业大学信息科学与工程学院,沈阳110870

出　　处：《无损探伤》2025年第2期11-13,19,共4页Nondestructive Testing Technology

摘　　要：针对传统神经网络模型在管道缺陷量化上存在较大误差的问题。本文提出基于ResNet改进的神经网络模型,融入漏磁检测理论,直接处理原始磁信号数据,提升缺陷量化准确性,在管道缺陷量化上取得了较好的效果。The traditional neural network model has a large error in the quantification of pipeline defects.In this paper,a neural network model based on improved ResNet,integrated with the theory of magnetic flux leakage detection,directly processes the original magnetic signal data,improves the accuracy of defect quantification,and achieves satisfactory results in pipeline defect quantification.

关 键 词：缺陷量化 ResNet 漏磁检测 原始磁信号 

分 类 号：TG115.28[金属学及工艺—物理冶金]