Real-coded genetic algorithm with oriented evolution towards promising region for parameter optimization
ZhiqiangChen1, Yun Jiang1, Xudong Chen2
COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12A) 93-101
1School of Computer Science and Information Engineering, Chongqing Technology and Business University, Chongqing, 400067, China
2Chongqing Engineering Laboratory for Detection Control and Integrated System, 19, Xuefu Avenue, Nan'an, Chongqing, 400067, China
In this paper, a novel real-coded genetic algorithm is presented to generate offspring towards a promising polygon field with k+1 vertex, which represents a set of promising points in the entire population at a particular generation. A set of 19 test problems available in the global parameter optimization literature is used to test the performance of the proposed real-coded genetic algorithms. Several performance comparisons with five significant real-coded genetic algorithms, three state-of-the-art differential evolution algorithms and three others significant evolutionary computing techniques are performed. The comparative study shows the proposed approach is statistically significantly better than or at least comparable to twelve significant evolutionary computing techniques over a test suite of 19 benchmark functions