TA679 : Reliability baxsed Optimization of Structures Using Game Theory Approach
Thesis > Central Library of Shahrood University > Civil & Architectural Engineering > PhD > 2022
Authors:
[Author], Ali Keyhani[Supervisor]
Abstarct: In multiobjective optimization problems, there are often conflicting objectives, finding an ideal state in which all objective functions are at their best at the same time is impossible. To solve this conflict, various methods such as Weighted sum, Goal Programming, ε-Constrainet, etc. have been proposed, which lead to finding the Pareto front. This front as a boundary between possible and impossible points is not an acceptable solution for engineering problems due to the lack of introduction of a unique solution. In this thesis, to resolve this conflict, it is proposed to present the concept of balancing and replacing it with the concept of optimization. By finding the equilibrium state, answers to the problem are searched that attract the attention of all the objective functions so that none of them have an incentive to violate it. Mathematicians have invented game theory since about a century ago to solve problems in which actors have conflicting interests and make decisions baxsed solely on their profits. In this branch of mathematical science, the result of each actor's decision is predicted baxsed on the interaction of others' answers. In the past few decades, various applications of it have been developed in economics, political, military and social sciences, and in this thesis, it has been used to solve structural engineering problems as interdisciplinary research. In this way, structures with the following conflicting objectives are selected: weight minimization (as an indicator for structure price), failure probability minimization (as an indicator for reliability) and displacement minimization (as an indicator for performance criteria). Each of the objective functions is assumed as the plaxyers of a game and by defining the Nash product and maximizing it, a state that has Nash equilibrium conditions is obtained. The findings of this research show that the reliability of the structure can be significantly increased with a small increase in the weight of the structure. Also, in the space of objective functions baxsed on the game theory approach, a balanced response has been obtained between the plaxyers minimizing the weight (lowest cost), minimizing the displacement of the top of the truss (maximum efficiency) and minimizing the probability of failure (maximum reliability). In addition, regarding the choice of materials used to build a skeleton structure (concrete or steel) when plaxyers want to make a decision with the criteria of minimum construction price and maximum reliability of the structure, a balanced answer has been found using game theory and with the help of the strategic form of the game. And it has been shown that choosing a concrete frxame and carrying out the design using the ultimate load and resistance method is a balanced answer to this problem. Also, as an example for using mexta-heuristic methods, with the help of NSGA II method, the Pareto front for a 25-bar truss has been obtained for specific values of the coefficient of variation for load and resistance. Keywords: reliability-baxsed optimization, failure probability of structures, game theory, Nash equilibrium, probability of failure of structures, multi-objective optimization, genetic algorithm, non-dominant sorting genetic algorithm.  
Keywords:
#Keywords: Reliability-baxsed optimization #Probability failure of structurs #Game theory #Nash equilibrium #Multi-objective optimization #NSGA II. Keeping place: Central Library of Shahrood University
Visitor: