Distortion Measurement of Ship Blocks Using Photogrammetry

Abstract

Welding being a transient phenomenon, explicit finite element thermomechanical analysis of large structural units is computationally extremely time-consuming and in general not feasible. Computational methods are available; however, they need to be validated with full-scale shop floor measured data. As of now, it is extremely difficult to do measurements and acquire some tangible data in actual shop floor situation. In the present work, a method was proposed to actually get the dimensions of such large structures through the use of multiple digital photographs of the structure. The photographs were then processed through a software to get to the dimensions of the threedimensional structure. For validating this method of measurement, two test cases were undertaken. First was using a flat surface plate measuring 3.3 m ×1.3 m over which black dots were pasted at predetermined positions. The distances between the dots were measured using a laser distance meter and compared with those obtained through photogrammetry. In the second case, a test sample was fabricated using a 500 m × 500 mm, 8-mm thick mild steel plate and two longitudinal and one transverse stiffener of 75 × 50 × 6 angle sections. Photographs were taken both before and after welding of the stiffeners. This set of photographs was processed to obtain the resulting deformation of the plate resulting from welding of the stiffeners. The distortion obtained through photogrammetry was compared with the distortion shape of the welded panel obtained from finite element analysis using an average plastic strain method. In both cases, the results of photogrammetry compared fairly well with those of laser distance meter measurements and finite element analysis. The proposed method shows great promise toward actual measurement of large blocks of ship structures and thereby satisfactorily validating theoretical methods of distortion prediction of such large structures.