Search Results

The complexities involved in an optimization problem at a system level require knowledge base that has information on different approaches and customization of these approaches to a specific class of the optimization problems. One approach that is commonly used is the metamodel based design optimization. The metamodel is 1) a conceptual model for capturing, in abstract terms, essential characteristics of a given optimization problem, and 2) a schema of sufficient formality to enable the problem modeled to be serialized to statements in a concrete optimization language [1]. Optimization is performed based on this metamodel. This metamodel approach has been proven effective and accurate in providing the global optimum. Depending upon the computational hardware availability in an organization, the metamodel based optimization could be much faster way of achieving the optimized solution. However, the accuracy of the optimization is highly dependent on the quality of metamodel generated.

The availability of computational resources has enabled an increased utilization of Design of Experiments (DoE) and metamodeling (response surface generation) for large-scale optimization problems. Despite algorithmic advances however, the analysis of systems such as water jackets of an automotive engine, can be computationally demanding in part due to the required accuracy of metamodels. Because the metamodels may have many inputs, their accuracy depends on the number of training points and how well they cover the entire design (input) space. For this reason, the space-filling properties of the DoE are very important. This paper utilizes a new group-based DoE algorithm with space-filling groups of points to construct a metamodel. Points are added sequentially so that the space-filling properties of the entire group of points is preserved. The addition of points is continuous until a specified metamodel accuracy is met.