Multi-objective Design Optimization of Engineering Systems: Uncertainty Approach and Practical Applications.

The increase in complexity of optimization problems results in an emerging need for simpler, faster and non-classical solutions. One of the options is conversion of a traditional non-hierarchical optimization system to a hierarchical system using an approach called multi-level (ML) decomposition (for optimization). Most of the work in the literature deals with the application of multi-level approach to deterministic optimization problems. But, in nature, many applications are uncertain, and hence, it is realistic to introduce uncertainty in the analysis and optimization. The first part of the present research deals with the development of a multi-level optimization procedure for uncertain engineering systems. The uncertainty in the problem is assumed to be stochastic and interval in nature. The methodology developed is illustrated by considering the optimization of structural and mechanical engineering problems. The second part of the present study deals in modifying a relatively new swarm intelligence technique based on the foraging behavior of ants called Ant Colony Optimization (ACO). A new multi-objective ant colony optimization algorithm is developed and applied to structural and mechanical engineering problems. The illustrative examples in the present research include the design optimization of an electric transmission tower (space truss), plane truss, gear box and the combustion chamber of an internal combustion engine. The third part of the research attempts to apply optimization techniques to practical engineering systems in the field of Heating Ventilation and Air Conditioning (HVAC) and Micro-Electronics. Novel design optimization models are created and hybrid optimization algorithms are developed for chiller plants and micro-channel heat exchangers used in electronic cooling. Illustrative case studies are performed.