Application of Array Magnetic Flux Leakage Detection System in Seamless Steel Pipe Inspection

Literature Overview

This paper, published in Physical Testing (2025, Vol. 43, No. 4), authored by Fan Qinghui, Zhou Changzhong, and Wang Wei from Baoshan Iron and Steel Co., Ltd. (Baosteel) Steel Pipe and Strip Division, presents an innovative array magnetic flux leakage (AMFL) detection system for seamless steel pipe quality control. The work addresses a well-recognized limitation of conventional magnetic flux leakage (MFL) inspection: the inadequate detection capability for short and oblique defects. The study provides both hardware architecture details and experimental validation through comparative testing on artificially notched sample pipes.

System Architecture and Technical Principles

The AMFL system described in this paper employs a multi-dimensional signal acquisition and fusion processing approach. The hardware architecture comprises four key subsystems: a lateral MFL system, a longitudinal MFL system, a three-roller centering and conveying device, and a host mounting platform. This configuration enables simultaneous detection of both longitudinal and transverse defects, which is a significant advancement over conventional single-orientation MFL systems.

Detection Performance Summary

The system's ability to achieve 100% detection rates for defects as short as 12 mm in both longitudinal and transverse orientations, and for oblique defects as short as 25 mm, represents a substantial improvement in inspection sensitivity. The signal-to-noise ratio (SNR) and repeatability were also evaluated, confirming the system's reliability for industrial deployment.

Engineering Significance for Seamless Pipe Manufacturing

From a manufacturing quality control perspective, this AMFL system addresses several critical challenges in seamless pipe production. Seamless pipes are subjected to a series of complex thermomechanical processing steps—including piercing, rolling, heat treatment, and cold drawing—each of which can introduce internal and external defects such as cracks, laps, seams, and inclusions. The conventional MFL method, while effective for larger defects, often misses short cracks that can initiate fatigue failure under cyclic loading, particularly in applications such as oil country tubular goods (OCTG) and pressure vessel piping.

The three-roller centering and conveying device is particularly noteworthy from an engineering standpoint. Proper centering is critical for MFL inspection because the magnetic flux distribution depends on the pipe's position relative to the magnetic field source and sensors. Any eccentricity introduces noise and reduces detection sensitivity. The three-roller design provides stable centering while allowing continuous pipe movement, which is essential for maintaining throughput in a production line.

Signal Processing and Defect Recognition

The multi-dimensional signal acquisition and fusion processing approach is the core innovation of this system. In conventional MFL inspection, the signal from a single sensor orientation is used to identify defects, which limits the ability to distinguish between different defect types and orientations. The AMFL system's simultaneous lateral and longitudinal MFL acquisition enables cross-orientation correlation of signals, which enhances defect recognition accuracy.

In my experience with MFL-based inspection systems, the primary challenge is distinguishing true defects from noise sources such as manufacturing marks, scale, and geometric variations. The array approach described in this paper likely employs signal correlation and pattern recognition techniques to filter out false indications. The reported improvements in SNR and repeatability support this interpretation.

Practical Considerations for Industrial Deployment

Several practical factors must be considered when deploying such a system in a seamless pipe production line:

• Throughput requirements: The system must maintain a pipe speed compatible with the production line's throughput, which depends on pipe diameter, wall thickness, and heat treatment cycle times.
• Pipe condition: Surface cleanliness, residual scale, and magnetic permeability variations can affect detection performance. A pre-cleaning step (pickling or shot blasting) may be necessary for optimal results.
• Calibration and verification: Regular calibration using reference notches of known dimensions is essential to maintain detection accuracy over time, as sensor degradation and magnetic field drift can occur.
• Data management: The multi-dimensional signal data generated by the array system requires robust data storage and processing infrastructure to support traceability and quality documentation.

Study Insights and Outlook

This study demonstrates the practical viability of array MFL technology for seamless pipe inspection, with quantifiable improvements in detection sensitivity for short and oblique defects. The 100% detection rate for defects ≥ 12 mm in length is particularly significant because such defects, while small, can serve as initiation sites for fatigue cracks in critical applications such as high-pressure gas pipelines and offshore oil and gas structures.

The work by the Baosteel team represents a meaningful contribution to the field of non-destructive testing (NDT) for steel pipes. The integration of multi-orientation MFL with advanced signal processing addresses a long-standing gap in seamless pipe quality control. For engineering practice, I would recommend that seamless pipe manufacturers evaluate the AMFL system as a complementary technology to existing inspection methods (such as ultrasonic testing and eddy current inspection), particularly for applications where short crack detection is critical.

Future research should focus on extending the detection capability to even shorter defects (below 12 mm), improving the system's ability to detect internal defects (such as internal cracks and laminations), and developing automated defect classification algorithms that can distinguish between different defect types based on the multi-dimensional signal patterns. Additionally, the system's performance under varying pipe conditions—different grades, wall thicknesses, and surface treatments—should be systematically characterized to establish clear operational guidelines.

In summary, this study presents a technically sound and practically valuable advancement in seamless pipe inspection technology, offering enhanced detection capability for short and oblique defects that directly contributes to improving the safety and reliability of seamless steel pipe products in demanding industrial applications.