Multi-objective optimisation of rolling rod product design using meta-modelling approach
Victor Oduguwa, Rajkumar Roy · 2002
Traditional solution methods such as search and sort for optimising complex real life engineering problems can be very expensive in terms of computational time. The considerable execution time tends to inhibit elaborate exploration of the design space and often results to sub-optimal solutions. This paper reports on an engineering optimisation approach designed to bridge the gap between traditional solution methods in the industry and state-of-the-art techniques from the research community. A modelling and optimisation technique has been developed using Design of Experiment (DoE) and meta-modelling approach to approximate expensive finite element (FE) runs. An evolutionary computational technique (NSGAII) is used for solving the optimisation problem. This solution technique was applied for multi-objective optimisation of a rod rolling design problem. The results showed NSGAII converge to the Pareto optimal front. The multiple optimal solutions help the designer in delivering a variety of optimal designs.