標題: Hybrid Genetic Algorithm with Mixed Mutation Mechanism for Optimal Display Panel Circuit Design
作者: Lo, I-Hsiu
Li, Yiming
Lee, Kuo-Fu
電信工程研究所
Institute of Communications Engineering
關鍵字: Differential Evolution;Genetic Algorithm;Panel Driver Circuit;Local Search
公開日期: 1-一月-2010
摘要: In this paper, a hybrid genetic algorithm which combines crossover operator of conventional genetic algorithm (GA) and mixed mutation mechanisms is proposed for the design and optimization of the panel gate driver circuits with amorphous silicon thin-film transistor (ASG driver circuit). The shooting local search method is applied on the first half of sorted population while the mutation of differential evolution (DE) is applied on the others after the end of crossover. This mechanism can accelerate the speed of convergence and keep the diversity in the meanwhile. For sizing of topology -given ASG driver circuits, the system which integrates proposed algorithm and circuit simulator to optimize crucial characteristics with practical constraints is implemented and tested. The results indicate that the proposed algorithm is effective and efficient, compared with conventional GA.
URI: http://dx.doi.org/10.1109/TAAI.2010.45
http://hdl.handle.net/11536/146516
ISSN: 2376-6816
DOI: 10.1109/TAAI.2010.45
期刊: INTERNATIONAL CONFERENCE ON TECHNOLOGIES AND APPLICATIONS OF ARTIFICIAL INTELLIGENCE (TAAI 2010)
起始頁: 222
結束頁: 225
顯示於類別:會議論文