Finite Element Analysis


Finite Element Analysis (FEA): 74%
Skills Devevloped: Finite Element Analysis (ABAQUS), Result Validations, Mechanics, Report Writing.
3rd Year Mechanical Engineering
2023



The aim of this experiment was to perform computational finite element analyses, determining the maximum stresses on a rectangular plate (made of 2000 series aluminium alloy) with a hole in the middle, while tensile forces were loaded on each end of the plate; as well as to develop an understanding of how mesh densities affect the accuracy of stress distributions, compared to theoretical results obtained by analytical stress concentration calculations with Equation: πœŽπ‘šπ‘Žπ‘₯ = π‘˜π‘‘ Γ— πœŽπ‘›π‘œπ‘š . Where Οƒmax is the maximum stress, ktis the stress concentration factor, and Οƒnom is the nominal stress.





Or

Scroll for Project Overview





BRIEF OVERVIEW



Objectives



To achieve the aims, the objectives are to:
  • Perform computational FEA on the geometry in Figure 1 with appropriate boundary conditions, tensile loads of 8000 N, and material properties with assumptions that yield strength is 324 MPa, ultimate tensile strength as 469 MPA at engineering strain of 19%, Young’s modulus as 73.1 GPa, shear modulus as 28 GPa, and Poisson’s ratio being 0.33.Β 
  • Employ suitable simplifications, assumptions, and approximations to the model for ideal efficiencies for computational modelling.Β 
  • Carry out simulations with different numbers of continuum (solid) type elements in the ranges of 30-60, 160-300, 600-1200, 2500-6000, and around 9000. Then, plot graphs of maximum stresses in the plate against the number of elements.Β 
  • Validate the modelled results compared with theoretical results obtained by Equation πœŽπ‘šπ‘Žπ‘₯ = π‘˜π‘‘ Γ— πœŽπ‘›π‘œπ‘š.Β 
  • Decide on the most appropriate method and mesh density with considerations of the accuracy, computational power, and efficiency of the experimental results, and formulate an analysis on the same structure with 20000 N loads and observe the differences.Β 


Figure 1: Plate with a central hole



Due to the symmetry of the shape, discretization method was used to model a quarter of the original plate .This allowed the ideal boundary conditions to be introduced, which prevented degrees of freedom that could complicate the analysis. Furthermore, the reduction of model size could possess a greater number of smaller meshing elements deployed, which could increase the accuracy of the results at where maximum stress occurred.

Results of simulation with element numbers of (1) 50, (2) 234, (3) 1128, (4) 4581, (5) 9087; the maximum stress on the hole were presented on the figures.



The experiment was conducted to analyse the effect of mesh densities on the accuracy of the modelled stress for a plate with a tensile load of 8000 N. It was learnt that as more element numbers were implemented, despite of greater computational time required, the results were more accurately demonstrated compared to the calculated theoretical results. Yet, when higher load was implemented on the same mesh density, the change in accuracy remained while the computational time increased significantly. Although results were differed to the theoretical results even in the highest tested element number, it was believed to be prevented by using more mesh elements, creating partitions, and using quadratic elements to improve the mesh controls for obtaining results of higher accuracy.





1st Year Truss Design


Integrated Design Project 1A: a part of 75%
Skills Devevloped: CAD (SolidWorks), Structural Analysis, Material and Cost breakdown, .
1st Year Mechanical Engineering
2020



The model was designed using two fink trusses connected horizontally. This design was selected as it is commonly used in roof structures which must withstand top-down loads similar to the given loading condition. It also reduced the complexity leading to less connections, keeping the cost under budget. Additional members were added to the centre of each fink truss and an extra triangle was created to reinforce the top beam upon which the force was applied for reinforcement. A simplified model of the fink truss was created using the SolidWorks weldments feature to model the truss with beams being joined together at pin joints. The beams were assigned with the material Balsa wood but with adjusted material properties of jelutong wood such as young’s modulus. Fixed geometry fixtures were applied at two points to each of the bottom horizontal members. They were placed 200mm from either side of the centre of the beam and restricted horizontal and vertical movement of those points. The load was applied across the upper most horizontal beam as a distributed load totalling 1000 N. One can see from the displacement image that it has been evenly distributed throughout the model. The most deflection occurs at the centre, but this was minimised with the reinforcing members to only around 0.34 mm.








Contact and Links

Email
LinkedIn
Portfolio PDF
Resume/CV
Note

I am completing my final year at Imperial College London as a master’s student in Sep 2025, and I am now looking for a job in the engineering sector to begin in 2025. Please reach out if you would like to know more about me, or just to connect and discuss more in an online coffee chat!
Β© 2020-2024 Ching Hang Ng, All Rights Reserved