Continuous and digitized monitoring and automated inspection are key parts of modern manufacturing and sustainment of aging infrastructure. The growing demand for these needs and shortage of required skill sets can slow down the global economy by increasing the risk or costs associated with catastrophic events. The diversity of requirements and specialized standards and codes around the world, along with the time-sensitive aspect of such inspections, makes automated fault detection and classification a prime application for utilizing artificial intelligence as an assistive tool that not only automates repeated tasks but also provides users with supporting inference to increase confidence before and during the inspection operation. In most critical cases, non-destructive testing (NDT) must be done once immediately after the weld is created and then on a scheduled or unscheduled repeated basis as the weld ages. One of the most commonly used NDT methods is radiography imaging using penetrating gamma or X-ray radiation. Existing assisted defect recognition tools in the literature are heavily focused on high-quality X-ray images and laboratory-focused imaging parameters, which in many cases are not representative of imaging done in real-world applications. Moreover, the literature has focused on welds that have undergone aging and have very clear defects—problems that classical image processing could easily address. This chapter and demonstration reviews the application of deep learning to find defects in newly created welds with minimal defect size and field-quality manually done welds as opposed to laboratory welds and addresses the industry standards for the classification of discontinuities (defects). First, the work developed and contextualized more than 100,000 X-ray images from various welds and annotated them with a group of NDT experts with varying years of experience. Based on this data and annotations, an optimized convolutional neural network (CNN) was designed and trained for detecting discontinuity and defects. Performance of the designed CNN was tested against other CNN architectures, and the overall accuracy of 96% overall classes was achieved.

Defect detection and classification in welding using deep learning and digital radiography

In reducing the costs associated with the fabrication of new pipelines for transporting hydrocarbons (pipelines), circumferential joint welding activities have a significant impact on total cost per km. LASER–MIG hybrid welding (LB–GMAW), with fibre-optic LASER beam delivery, offers the opportunity for reducing fabrication costs, thanks to increased productivity, despite the higher initial investment. The aim of this work is to demonstrate the feasibility of pipeline fabrication using LB–GMAW hybrid welding technology for the circumferential welding of large diameter pipes (36″) for gas pipelines, on X100 high-grade steel with nominal thickness of 16 mm. Solid-state LASER systems with fibre-delivered LASER beam (Nd:YAG and Yb:SiO2) have been used in this study. The hybrid welding system has been shown to be of potential interest since, by guaranteeing adequate mechanical properties, it ensures increased productivity (reduced number of passes and increased welding speed).

Circumferential welding of gas pipeline pipes using hybrid technology with fibre-delivered LASER beamPaper presented at the 4th National Welding Day, Workshop: “New frontiers in LASER beam welding, friction stir welding and electron beam welding processes” Genova, 25–26 October 2007.

During cross-country underground pipeline construction welding is one of the major construction activity. At various parts of the world, many pipeline welding machines are in use to produce defect-free economic pipeline joints. Every weld machine manufacturer claims its own advantages and other’s drawbacks. Two famous welding machine manufacturing companies for pipelines are PWT (Italy) and CRC-Evans (USA). Thus, the current work at the cross-country gas pipeline construction site from Vijaypur (MP, India) to Dadri (UP, India) (VDPL) is taken up to determine which of the above two mentioned machines are technically better and relatively cheaper while determining weld joint cost and overall pipeline construction cost. Data of two construction sites of VDPL namely Guna and Mathura are collected for weld joint construction and repair. The sites were constructed by ESSAR and EIL for GAIL. The carbon steel pipeline being laid was of 48” nominal size. A detailed cost analysis of the machines is estimated by considering all the related man, machine and material cost, both for mainline weld and weld joint repair. Technical performance of the weld machines is evaluated by comparing various weld parameters for pipeline weld joints. Performance of both of the machines is found to be comparable and almost equal in terms of total cost involved but the repair cost of PWT is less than CRC. Both PWT and CRC are found to produce weld defects with defect of incomplete fusion (IF) more frequently. Suitable remedies are suggested for improvement.

Onsite Technical and Economic Performance Evaluation of PWT (Pipeline Welding Technology): A Comparative Analysis with CRC-Evans Welding Technology

Kljucne reci: Cevi, cevovodi, projekti cevovoda, termomehanicki proces valjanja, celici za cevovode X100, X120, razliciti postupci zavarivanja, rucno elektrolucno zavarivanje obloenom elektrodom, mehanizovano zavarivanje, CAPS, mehanicke osobine, napon tecenja, zatezna cvrstoca, tvrdoca i energija udara.

/pdf/celici-za-cevi-x100-i-x120-the-pipeline-steel-x100-and-x120-1tx6e1cpeo.pdf

Čelici za cevi x100 i x120 the pipeline steel x100 and x120

The Oil and Gas industry is constantly looking for ways to reduce capital investment for new projects while maintaining performance and quality standards. As the demand for oil and natural gas increases, so does the need for an expanded network of onshore and offshore transmission pipelines. However, the capital investment associated with new pipelines is enormous. There are a number of materials joining technologies that are being investigated to reduce pipeline construction costs. This document presents three technologies showing particular promise: laser pipeline welding, the use of high strength pipeline materials, and advanced automatic inspection. Test have shown that the HLAW process, which combines the LBW and GMAW processes, is felt to offer a number of benefits in terms of higher potential productivity, ability to produce welds with excellent material properties and a reasonable tolerance for fit-up variations. For existing laser power levels, a viable scenario for implementation of lasers for pipeline welding applications is the use of HLAW for root bead welding, followed by joint filling with conventional mechanised GMAW-P. Concerning the use of high strength pipeline materials, results of a project are presented. The primary objective of this study was to evaluate the mechanical properties of weld metals deposited using a number of commercially available welding consumables and welding processes. This work included characterization of weld metal microstructure, HAC susceptibility of the GMA weld metal and X100 base metal, and solidification cracking of selected welding consumables. The project evaluated commercially available consumables using several welding processes and shielding gases. The third part of the document covers results of a project aiming to collect third-party independent data and statistically characterise the systematic and random errors in girth weld defect sizing, as measured by mechanised ultrasonic testing P/E, TOFD methods and PA ultrasonic technology, in support of pipeline reliability assessments. The defect sizing accuracy of the automated ultrasonic testing (AUT) of girth welds was evaluated using zonal discrimination with focused and non-focused search units arranged for P/E and TOFD methods and PA technology. The objectives were achieved by conducting “open” and “blind” UT trials. NDE flaw verification and destructive testing for flaw size validation were performed. AUT data were collected and statistically analysed.

M. Johnson

Papers

Reducing pipeline construction costs: New technologies