A New Framework with FDPP-LX Crossover for Real-Coded Genetic Algorithm

作者：Chen Zhi Qiang^*; Wang Rong Long

来源：IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences, 2011, E94A(6): 1417-1425.

DOI：10.1587/transfun.E94.A.1417

• genetic algorithm
• function optimization
• real-coded
• crossover operator

This paper presents a new and robust framework for real-coded genetic algorithm, called real-code conditional genetic algorithm (rc-CGA). The most important characteristic of the proposed rc-CGA is the implicit self-adaptive feature of the crossover and mutation mechanism. Besides, a new crossover operator with laplace distribution following a few promising descent directions (FPDD-LX) is proposed for the rc-CGA. The proposed genetic algorithm (t-c-CGA+FPDD-LX) is tested using 31 benchmark functions and compared with four existing algorithms. The simulation results show excellent performance of the proposed rc-CGA+FPDD-LX for continuous function optimization.

• 出版日期2011-6