Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/487213
Title: | Non-probabilistic method on edge crack plane with presence of fuzzy variables |
Authors: | Yusmye Nur Abu Yusuf (P60191) |
Supervisor: | Ahmad Kamal Ariffin Mohd Ihsan, Prof. Ir. Dr. |
Keywords: | Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia Crack structure Structural analysis Fracture mechanism |
Issue Date: | 13-Nov-2021 |
Description: | Fracture mechanism is a main approach to enable the analysis of the cracked structures. However, in the structural analysis, the presence of uncertainty becomes challenging that should be avoided to prevent a material failure. Typically, structural analysis depends on the system parameters such as material properties, geometry properties, material properties, boundary conditions and applied loads which are considered as deterministic input variables. Nevertheless, rather than these values, in engineering it may have insufficient, inadequate, indistinct and vagueness data or information about the variables used. The main objective of this research work is to develop the Fuzzy Finite Element Method (FuzzyFEM) for single edge crack caused by uncertainties. Thus, modeling these types of uncertainties becomes a vital concern. Many researchers model the uncertainty problem through a probabilistic method. Unfortunately, reliable results with adequate precision is a big challenge with limited experimental data. In recent decades, fuzzy theory and interval analysis are becoming powerful tools, in which these two methods are a part of the non-probabilistic method. Therefore, this thesis develops a method to determine the stress intensity factor (SIF) in crack edge problem by combining the finite element methods (FEM) with fuzzy set theory to become a fuzzy finite element method (FuzzyFEM) for Mode I and II. The selected uncertain variables are represented in the form of a fuzzy number and modeled by a membership function while applying this approach. The modeling of the triangular membership function in early stage is a part of the fuzzification process. The next stage is the important process to implement the deterministic FEM, which is known as the mapping process. In this stage, the fuzzy input is map through fuzzy arithmetic rule to obtain the fuzzy output. The vertex method is a numerical method for fuzzy arithmetic that combines the interval analysis method with-cut methods. By using interval analysis method, the maximum and minimum fuzzy values are determined for each-level and carried out the mapping process. The result from the mapping process is the fuzzy output for each-level too. The determination of stress intensity factors (SIF) is a final output for this research work. Before the crisp SIF gained the values of output are in a fuzzy form, so the defuzzification processes need to be applied. Defuzzification is the process to transform the fuzzy output to crisp output by using the center of gravity method. As a result, the developed FuzzyFEM model has resolved the uncertainties problem in fracture analysis efficiently in terms of time and cost. Fuzzified the two variables shown the small error, and these show the best agreement on SIF output. While fuzzified four variables shown best fit agreement with the six-sigma concept. This means that the more fuzzy input is taken into account, the more uncertainty is considered, and the width of the graph for the fuzzy output membership function is getting large. Thus, the output will be more conservative and precise. The obtained solutions are depicted in terms of figures and tables to show the efficiency and reliability of the present analysis.,Ph.D. |
Pages: | 155 |
Publisher: | UKM, Bangi |
Appears in Collections: | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
ukmvital_125317+SOURCE1+SOURCE1.0.PDF Restricted Access | 2.78 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.