An Accuracy Assessment Method for Two-Dimensional Functional Data in Simulation-Based Design

Accuracy assessment methods are necessary for functional data used in simulation-based design optimization to ensure model and optimal solution validity. Although many error metrics exist and perform well for one-dimensional (1D) applications, the suitability of such metrics for higher dimensional functional data, such as two-dimensional (2D) performance maps, have been largely unexplored. This paper examines the extension of the 1D accuracy assessment method AVASIM to 2D applications, in support of decomposition-based multidisciplinary design optimization (MDO) coordination strategies that require measurement of the consistency of functional data exchanged among subproblems through error metrics. Specifically, AVASIM is used as a consistency measure in the coordination strategy of analytical target cascading that requires functional data consisting of 1D torque curves and 2D power loss maps from motors to be exchanged among subproblems in electric vehicle design optimization. Results indicate that a generalized AVASIM formulation is an effective consistency measure for accuracy and consistency in a computationally efficient manner.