一种管道环焊缝缺陷识别方法以及装置
By converting pipeline voltage data into grayscale images and training the model, the errors and inefficiencies of manual interpretation in magnetic flux leakage detection are solved, achieving efficient and accurate identification of pipeline circumferential weld defects and supporting intelligent pipeline inspection.
CN118088943BActive Publication Date: 2026-07-17PIPECHINA SOUTH CHINA CO +2
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- PIPECHINA SOUTH CHINA CO
- Filing Date
- 2024-01-05
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing magnetic flux leakage detection technology relies on manual interpretation for identifying defects in pipeline circumferential welds, which results in large labeling errors, low efficiency, and significant differences in personnel experience, leading to high uncertainty.
Method used
The pipeline voltage data is converted into raw grayscale images, preprocessed, and divided into training, validation, and test sets. A training model is constructed and analyzed using a target recognition model to identify defects in the pipeline circumferential weld.
Benefits of technology
It improves the accuracy and efficiency of identifying defects in pipeline circumferential welds and supports intelligent pipeline inspection.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN118088943B_ABST
Abstract
本发明提供一种管道环焊缝缺陷识别方法以及装置，属于管道识别技术领域，方法包括：从检测器中获得管道电压数据，并将管道电压数据转化为原始管道灰度图；对原始管道灰度图进行预处理，得到待处理管道图像数据集；将待处理管道图像数据集划分为管道图像训练集、管道图像验证集以及管道图像测试集；构建训练模型，通过管道图像训练集和管道图像验证集对训练模型进行模型改进得到目标识别模型；通过目标识别模型对管道图像测试集进行识别分析得到管道环焊缝缺陷的识别结果。本发明有效地提高了管道环焊缝缺陷的识别效果，极大地提高了数据处理的效率和准确度，对于实际漏磁信号识别以及管道智能化检测具有重要的作用。
Need to check novelty before this filing date? Find Prior Art