Use of Synthetic Flaws to Assess Pipeline Seam Weld Inspection Performance

Pipeline seam welds are often inspected using ultrasonic In-Line Inspection (ILI) technologies. The measurement performance specification of an ultrasonic ILI tool is based on simple, planar, machined notches which are very reproducible, but are not representative of the complex flaw morphologies that occur naturally in seams such as hook cracks and tilted lack of fusion flaws. In order to assess ILI performance on naturally occurring flaws, “in-the-ditch” Nondestructive Testing (ITD NDT) is performed to validate a subset of the population of ILI reported features. Due to the limited number, type, and dimensional (height and length) uncertainty of these flaws, the field validation approach has limitations in terms of efficiency and accuracy in determining ILI detection capabilities and sizing performance. Recently, specialized synthetic flaw fabrication technology has been developed and provides complex, natural crack-like morphologies with reliable and reproducible size dimensions. Effective validation spools with flaws (of representative geometries) can be achieved through engineered designs that consider the number, size and shape of manufactured flaws. This enables owners to quickly and reliably assess the performance of both ILI tools and ITD NDT operators. Assessing performance with the synthetic flaw approach provides results that are more comprehensive and cost-effective compared to the typical field validation approach alone. This is because the flaw population is designed rather than randomly selected from excavation data. This paper addresses the design, use and field experience with validation spools. This paper will present the performance of ILI tools and UT examiners based on synthetic flaw qualification exams, and how this supports related ILI and operator validation work.

Spatial data analysis for laser scanning based shape error inspection

Laser scanning, a widely used technology, has been highly developed and adopted in various industrial applications. The methodologies used for scanner date processing are mostly point based. In this thesis, a new approach is presented to analyze spatial data obtained from a 3-D laser scanner for shape error inspection. Different from traditional methodologies, the method proposed in this research is frequency based. The method utilizes the Fourier transform to decompose a 2-D curve or 3-D shape into its spatial components by applying two 1-D FFT (Fast Fourier Transform) on 2-D curves or two 2-D FFT on 3-D shapes. The spatial components including frequency, amplitude, and phase are defined as shape characteristics to represent the shape under inspection. By relating spatial components with GD&T (Geometric Dimensioning and Tolerancing) standards using proper analysis techniques, such as frequency spectrum and cross correlation, shape errors can be detected and characterized. One of the applications of this method is automated inspection. In this research, the spatial data method is applied to MIG (Metal Inert Gas) weld inspection. Experiments are carried out to analyze the 2-D curve of a projection weld data, and the 3-D scanning data directly. A MIG weld inspection system is also developed for production use.

Spatial data analysis for laser scanning based shape error inspection

Laser scanning, a widely used technology, has been highly developed and adopted in various industrial applications. The methodologies used for scanner date processing are mostly point based. In this thesis, a new approach is presented to analyze spatial data obtained from a 3-D laser scanner for shape error inspection. Different from traditional methodologies, the method proposed in this research is frequency based. The method utilizes the Fourier transform to decompose a 2-D curve or 3-D shape into its spatial components by applying two 1-D FFT (Fast Fourier Transform) on 2-D curves or two 2-D FFT on 3-D shapes. The spatial components including frequency, amplitude, and phase are defined as shape characteristics to represent the shape under inspection. By relating spatial components with GD&T (Geometric Dimensioning and Tolerancing) standards using proper analysis techniques, such as frequency spectrum and cross correlation, shape errors can be detected and characterized. One of the applications of this method is automated inspection. In this research, the spatial data method is applied to MIG (Metal Inert Gas) weld inspection. Experiments are carried out to analyze the 2-D curve of a projection weld data, and the 3-D scanning data directly. A MIG weld inspection system is also developed for production use.