This thesis aims to analyze the effects of geometric imperfections on the load-bearing capacity of high-strength steel grades and how the industry implements these imperfections in Finite Element Method (FEM) modeling. The goal was to examine the industry's implementation of these geometric imperfections in relation to compliance with established standards and regulations. Through conducted interviews, hand calculations, and numerical simulations, the study provided SSAB with a deeper understanding of geometric imperfections. The interview focused on handling geometric imperfections in the manufacturing of truck cranes, exploring various strategies to ensure structural integrity and compliance with industry standards. The company in focus oversized the construction in nominal analyses and followed EN 13001 and internal guidelines to prevent the effects of imperfections. A test specimen from SSAB's laboratory was used as a reference against the simulations. The test specimen consisted of a high-strength steel profile mimicking those used in cranes. The geometry of the test specimen was then applied to the numerical simulations In numerical simulations, the flat and round sides of the test specimen were compared under compression. When the round part was in compression, the simulation underestimated the moment capacity by approximately 14 kN, equivalent to about 7.1%, compared to the actual test results. When the flat part was in compression, the simulation overestimated the moment capacity by approximately 7 kN, equivalent to 8.4%. The differences between simulations and tests were relatively small, and simulations were deemed quite representative compared to tests. Simulations with imperfections showed marginal effects on load-bearing capacity. For the profile simulated with imperfections, the load-bearing capacity before failure was 84.5 kN, while the capacity for the profile without imperfections was 82.6 kN, with a difference of 2.24%. No major conclusions regarding the impact of imperfections can be drawn with such a small difference, but it is interesting that the profile with applied imperfections has 2.24% better load-bearing capacity than the one without. The impact of thickness on load-bearing capacity was also examined. The most significant difference noted between the ideal geometry and the one with imperfections was at a thickness of 8 mm. The main reason imperfections made the most difference there is the slenderness. Thinner thicknesses of 2, 4, and 6 mm were so slender that all would be limited by local buckling. For the larger thickness of 10 mm, the idea was that the profile becomes thick and rigid enough to avoid buckling affecting load capacity. In the case of 8 mm, the cross-section was right on the border between cross-section class 3 and 4, where imperfections take a larger part of the cross-section to class 4. It can be concluded that in cases where the part in compression is right on the verge of being so slender that cross-section reduction is almost relevant, imperfections can significantly reduce load capacity. It is noted that thicker profiles can be affected by imperfections much more than slender ones. Hand calculations revealed differences between calculated and experimental failure loads, varying between 18% and 29%. These differences can be attributed to discrepancies in strength class and the geometry of the test component. Adjusting the strength class to 850 MPa in hand calculations improved the agreement with experiments. Geometric uncertainties include variations in thickness, where a larger thickness increases load capacity. Additional uncertainties arise for the flat part regarding cross-section reduction. In conclusion, hand calculations align reasonably well with test results, but differences were scattered and challenging to attribute to geometric imperfections. For future studies, a closer examination of the company's method with safety factors for imperfection calculations is suggested, along with investigations into cross-sectional profiles and the transition between cross-section classes. Furthermore, the need for more simulations with different geometries is emphasized to better understand the effects of geometric imperfections.
Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:ltu-107559 |
Date | January 2024 |
Creators | Karlsson, Marcus, Sjöström, William |
Publisher | Luleå tekniska universitet, Institutionen för samhällsbyggnad och naturresurser, Luleå tekniska universitet |
Source Sets | DiVA Archive at Upsalla University |
Language | Swedish |
Detected Language | English |
Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
Format | application/pdf |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.0023 seconds