| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 263.06 KB | Adobe PDF |
Advisor(s)
Abstract(s)
In view of the fact that Genetic Algorithms (GAs) are not well suited for fine-tuning structures that are close to optimal solutions [1], this paper suggests the incorporation of local improvement operators into the GA recombination phase. This study presents a hybrid genetic algorithm, also known as Memetic Algorithm (MA), applied to the design of combinational logic circuits. MAs are evolutionary algorithms (EAs) that apply a separate local search process to refine individuals (i.e. that improve their fitness by hill-climbing). Under different contexts and situations, MAs are also known as hybrid EAs or genetic local searchers. The proposed MA associates a GA with the gate type local search (GTLS). Combining global and local search is a strategy used by many successful global optimization approaches, and MAs have in fact been recognized as a powerful algorithmic paradigm for evolutionary computing. We also modify the calculation of the fitness function by including a discontinuity evaluation that measures the error variability of the Boolean table. The results show an improvement of the final fitness function followed by a reduction of the average number and the standard deviation of generations required to reach the solutions, for all the tested circuits.
Description
Keywords
Evolutionary algorithms Logic circuits Memetic algorithms