Reliability estimation for aluminum alloy welded joint with automatic image measurement of surface crack growth. Issue 4 (13th June 2016)

• Record Type: Journal Article
• Authors: Huang Peng, Qizi; Liu, Tianyu; Sun, Quan; Huang, Wenwei
• Is Part Of: Engineering computations. Volume 33:Issue 4(2016)
• Page Start: 1205
• Page End: 1223
• Publication Date: 2016-06-13
• Subjects: Image measurement -- Reliability estimation -- Surface crack growth; Computer-aided engineering -- Periodicals; Computer graphics -- Periodicals; 620.00285
• DOI: 10.1108/EC-04-2015-0092
• Languages: English
• ISSNs: 0264-4401

Abstract

Abstract : Purpose: – As an important connecting component, the reliability of aluminium alloy welded joints influences the whole structural effectiveness and stability of equipment. The purpose of this paper is to propose a novel reliability estimation approach to the welded joints based on time-transformed Wiener process with automatic image measurement of crack growth. The crack length information of the welded joints is incorporated into reliability analysis to reflect the product time-varying characteristics. Design/methodology/approach: – The proposed approach is superior to other crack growth estimations in that it innovatively introduce a non-contact and flexible photogrammetry technique.First, on-line crack growth images of aluminium alloy welded joints are acquired by the designed monitor system. Second, crack length is calculated with image measurement, then the crack growth data during the manufacturing process is prepared. Finally, a time-transformed Wiener process is used to modeling the degradation, and reliability estimation is carried out with Wiener model. The approach has been validated on five 7075-T7351 welded joint samples. Findings: – The method has a twofold task: first, the extraction of crack length growth data by a sequence of image processing. The main step is to model the crack skeleton with crack skeleton tree, and remove it edges to calculate the length of crack; second, the prediction of crack growth and reliability estimation. Research limitations/implications: – The limitation of proposed method should not be ignored. The pixel/mm scale should be calibrated in advance that means once we have built the monitor system, the relative position of the CCD camera and the surveyed crack cannot change anymore. It has reduced the flexibility. To improve this, we can obtain binocular vision in crack image measurement. The 3-D measurements could solve calibration problem and provide more information, such as the depth and the orientation of crack to research. Therefore, future work can be centered on the improvement of monitor system and measurement precision. Originality/value: – In the paper a novel method to estimate reliability of crack growth from welded joint based on image measurement has been presented. This method could be widely applied in different filed of manufacturing systems, reliability engineering and structural analysis.