The Research progress and prospect of data mining methods on corrosion prediction of oil and gas pipelines

Lamb Wave Tomography for Defect Localization using Wideband Dispersion Reversal Method

A novel methodology based on the reflected L(0,1) guided wave for quantitative detection of corrosion-induced wall thickness loss in continuous pipes

Explainable deep learning based ultrasonic guided wave pipe crack identification method

Currently, with the advancement of the urbanization, the pipeline explosions may lead to a very large catastrophe in the city. Pipeline networks are regularly inspected using nondestructive evaluation (NDE) methods such as smart pigs (cylinder-shaped electronic devices to detect metal loses), mapping tools based on GPS for above ground pipelines, guided wave ultrasonics and hydrostatic testing. The major challenge for inspecting pipeline networks occurs for unknown map of distribution, where geometric and spatial features may not be known, and the properties of pipe structure make pigging impossible. Therefore, this research proposed to identify the geometric and spatial signatures in the pipeline networks in urban settings by means of elastic wave propagation. Firstly, the geometric and spatial complexity was defined by extracting the features from the pipeline network, which includes thickness of wall, diameters, types of connections, path, materials and so on. Secondly, the influence of complexity on the propagation characteristics of ultrasonic waves in pipeline network is investigated by numerical simulation of different pipeline configurations. Then the sensitive features were selected from results obtained by parametric study. Therefore, the geometric distribution of improperly documented pipeline networks can be identified and then their structural state is assessed accordingly. 

Identification of Geometric and Spatial Complexities in Pipeline Based on Ultrasound

The converted modes generated by mode conversion in guided waves-based inspection can provide plenty of defect information, and the reasonable usage of multiple wave modes can improve the results of defect inspection and evaluation. Axisymmetric longitudinal L(0,2) mode guided-waves is widely used in pipeline inspection currently, and the lower order longitudinal L(0,1) mode is easily excited simultaneously. This paper focuses the interconversion process of axisymmetric modes for incident longitudinal modes when interacting with the defects in pipeline whist ignoring the converted non-axisymmetric modes through suppressing them in reception. This process is defined as symmetric mode conversion in this research. The pattern of axisymmetric mode sequence during symmetric mode conversion is identified and the method is proposed to extract the converted axisymmetric mode components for analysis. The relationships between the defect features and the modes generated in symmetric mode conversion under the excitation of longitudinal mode waves are investigated. The conclusion obtained by numerical simulation is also experimentally verified using guided wave data from practical pipeline. The results show that longitudinal modes in symmetric mode conversion present the useful time and frequency characteristics, which provides the potential for establishing an effective defect inspection and evaluation method with longitudinal guided waves.

Time-frequency characteristics of longitudinal modes in symmetric mode conversion for defect characterization in guided waves-based pipeline inspection

Heming Gao

Papers

Time-frequency characteristics of longitudinal modes in symmetric mode conversion for defect characterization in guided waves-based pipeline inspection