A Framework for Simulation-Based Optimization with Application to Green Building Design
Weimin Wang, Hugues Rivard, Radu Zmeureanu · 2005
This paper presents an object-oriented framework that gives attention to the features of simulation-based optimization problems such as hierarchical variables and the coupling with simulation programs. The framework consists of three basic modules: the variables module, the simulation module, and the optimizer module. The variables module defines variables and organizes them according to their relationships. The simulation module evaluates objective functions and functional constraints. The optimizer module implements optimization algorithms. The genetic algorithms implemented in the optimizer module can solve (1) unconstrained single objective optimization problems, (2) constrained single objective optimization problems, and (3) unconstrained multi-objective optimization problems. The application of this framework is demonstrated through a case study for green building design.