Probabilistic based topology optimization in ANSYS

Abstract

Optimization is a process or methodology of making something as fully perfect, functional or effective as possible. Structural optimization in the thesis focuses on optimal design of the loaded structures. The optimal topology design which is a type of structural optimization is defined as optimization problem of material distribution. The material densities of finite elements serve as variables and the maximum stiffness is defined as objective function. The interest for finding optimal design grows in recent years because it gives us a possibility to save the material that doesnt significantly contributes in the global stiffness of the structure. Probabilistic design is widespread discipline in engineering. It considers uncertainties in the structures. These uncertainties and random effects are related to quality and reliability of the structure. Probabilistic design is not used very often in the civil engineering because engineers rely on prescribed building codes and safety factors. In recent years there has been a growing interest to combine probabilistic design with topology optimization. Techniques for combining these two analyses presented to this day are very time consuming and therefore expensive. They are also unreachable to every engineer. In this work a simplified method for engineers is presented. It combines the deterministic load cases that came from probabilistic analysis into an end topological solution. Monte Carlo simulation is applied for the probabilistic analysis.

Optimization is a process or methodology of making something as fully perfect, functional or effective as possible. Structural optimization in the thesis focuses on optimal design of the loaded structures. The optimal topology design which is a type of structural optimization is defined as optimization problem of material distribution. The material densities of finite elements serve as variables and the maximum stiffness is defined as objective function. The interest for finding optimal design grows in recent years because it gives us a possibility to save the material that doesnt significantly contributes in the global stiffness of the structure. Probabilistic design is widespread discipline in engineering. It considers uncertainties in the structures. These uncertainties and random effects are related to quality and reliability of the structure. Probabilistic design is not used very often in the civil engineering because engineers rely on prescribed building codes and safety factors. In recent years there has been a growing interest to combine probabilistic design with topology optimization. Techniques for combining these two analyses presented to this day are very time consuming and therefore expensive. They are also unreachable to every engineer. In this work a simplified method for engineers is presented. It combines the deterministic load cases that came from probabilistic analysis into an end topological solution. Monte Carlo simulation is applied for the probabilistic analysis.