| Real world optimization problems are usually dynamic multiple objective in nature with several constraints on the function values in the presence of uncertainties and bounds on the inputs. This work proposes systematic segmentation of the said problem using Cultural Algorithms. This is achieved by maintaining feasible and infeasible best individuals and their fitnesses and constraint violations in the Situational Space, bounds for the search in the Normative Space, crowding information in the Topographic Space, and function sensitivity and relocation offsets in Historical knowledge space of the Cultural Algorithm. The information is used to vary the flight parameters of the PSO equations, to generate newer individuals and to track dynamic multiple objective optima with constraints. |
