Title :
Design exploration: extending the bounds of the search space
Author :
Beck, M.A. ; Parmee, I.C.
Author_Institution :
Eng. Design Centre, Univ. of Plymouth, UK
Abstract :
It is during conceptual design that neither the problem space nor solution space is fixed and the designers aim is an exploration of a range of possible problem-solution scenarios. Within such a framework the constraints placed upon the initial limits assigned to the design variables are subject to change. Thus this paper presents a multi-population GA approach that allows the search to move outside of its pre-defined limits. Utilising dynamic fitness sharing and directional search techniques sub-populations are encouraged to explore differing high performance and unspecified areas of the search space
Keywords :
CAD; adaptive systems; genetic algorithms; search problems; conceptual design; design exploration; design variables; directional search techniques; dynamic fitness sharing; multi-population genetic algorithm approach; problem space; problem-solution scenarios; solution space; sub-populations; Convergence; Design engineering; Genetic mutations; Process design; Programmable logic arrays; Steady-state; Wheels;
Conference_Titel :
Evolutionary Computation, 1999. CEC 99. Proceedings of the 1999 Congress on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5536-9
DOI :
10.1109/CEC.1999.781977