 |
International Society of Science and Applied Technologies |
| Reliability Assessment for Corroded Pipes Based on MFL Inspection | ||||
| Author | Yinuo Chen
|
|||
| Co-Author(s) | Zhigang Tian; Haotian Wei; Shaohua Dong
|
|||
| Abstract | The magnetic flux leakage (MFL) inspection is the most common method for detecting defects in corroded pipelines and assessing the reliability of pipelines. However, in traditional ways, many steps are required to obtain the pipeline's reliability from MFL detection. This study establishes a finite element (FE) model to simulate the MFL signal. Then, a novel method is proposed to use a convolutional neural network (CNN) to directly map the MFL image to the array representing the pipeline's reliability in the future 30 years. The case study demonstrates that the proposed method can estimate corroded pipe reliability more quickly and precisely by eliminating several image processing and calculation procedures. The results indicate that the proposed method is beneficial for assisting pipeline operators in efficiently identifying the reliability status and providing a basis for pipeline integrity management.
|
|||
| Keywords | Corroded pipeline; convolutional neural network; magnetic flux leakage; finite element model; reliability | |||
| Article #: RQD28-56 | ||||
Proceedings of 28th ISSAT International Conference on Reliability & Quality in Design |