Stochastic design optimization accounting for structural and distributional design variables

Purpose This paper aims to present a new method, named as augmented polynomial dimensional decomposition (PDD) method, for robust design optimization (RDO) and reliability-based design optimization (RBDO) subject to mixed design variables comprising both distributional and structural design variables. Design/methodology/approach The method involves a new augmented PDD of a high-dimensional stochastic response for statistical moments and reliability analyses; an integration of the augmented PDD, score functions, and finite-difference approximation for calculating the sensitivities of the first two moments and the failure probability with respect to distributional and structural design variables; and standard gradient-based optimization algorithms. Findings New closed-form formulae are presented for the design sensitivities of moments that are simultaneously determined along with the moments. A finite-difference approximation integrated with the embedded Monte Carlo simulation of the augmented PDD is put forward for design sensitivities of the failure probability. Originality/value In conjunction with the multi-point, single-step design process, the new method provides an efficient means to solve a general stochastic design problem entailing mixed design variables with a large design space. Numerical results, including a three-hole bracket design, indicate that the proposed methods provide accurate and computationally efficient sensitivity estimates and optimal solutions for RDO and RBDO problems.

The Application of Artificial Immune System in Mechanical Engineering for Constrained Optimization

A hybrid AIS-GA was proposed and tested. The algorithm performed very well in problems presenting continuous, discrete, and mixed design variables, producing feasible solutions in all runs for all problems considered. Also, it is much more easily parallelizable than the previous hybrids, and does not require any user-defined parameter other than the parameters already used by the AIS and the GA.

Particle Swarm Methodologies for Engineering Design Optimization

Particle swarm methodologies are presented for the solution of constrained mechanical and structural system optimization problems involving single or multiple objective functions with continuous or mixed design variables. The particle swarm optimization presented is a modified particle swarm optimization approach, with better computational efficiency and solution accuracy, is based on the use of dynamic maximum velocity function and bounce method. The constraints of the optimization problem are handled using a dynamic penalty function approach. To handle the discrete design variables, the closest discrete approach is used. Multiple objective functions are handled using a modified cooperative game theory approach. The applicability and computational efficiency of the proposed particle swarm optimization approach are demonstrated through illustrate examples involving single and multiple objectives as well as continuous and mixed design variables. The present methodology is expected to be useful for the solution of a variety of practical engineering design optimization problems.