Multi-disciplinary design Optimization based on metamodel

The real-world design optimization problems are complex and multidisciplinary. For example, design of a satellite system involves complex interactions between various sub-systems with a large number of design variables and constraints that limits application of conventional design optimization methods to this class of problems. In addition, design search space of such problems can be large and non-convex involving simulation of several interacting disciplines and hence it is time consuming or difficult to rapidly evaluate trade-offs between various disciplines. To address these difficulties, several research works are focused on the multidisciplinary design optimization methods. In this respect, this paper presents and efficient multidisciplinary design optimization framework using metamodels. The proposed method extends multidiscipline feasible concept using metamodels, design of experiments (DoE) and sequential quadratic programming (SQP) for solving large scale design optimization problems such as satellite systems. The proposed method is compared with the existing methods using a number of benchmark problems. The proposed method is implemented on a remote sensing satellite system. The results obtained show that the proposed method provides an effective way of solving large-scale design optimization problems.