A center-of-gravity-based recombination operator for genetic algorithms

A new recombination operator is introduced and treated in the paper. It performs specific crossover (breeding) between the two fittest parental chromosomes producing a new child chromosome, which is based on the center of gravity (CoG) paradigm. This new child chromosome is one of the members of the new population. The rest of the chromosomes are produced by the conventional procedures. The new operator could be used both in a binary as well as a real-coded GA.. The insight of the proposed mechanism as well as the test results indicate that it leads to better results in most cases or at least the same results as without its use. With almost no increasing computational expenses the speed of convergence as well as the final result in tests surpass the conventional approach. This new approach has been tested with a practical problem of scheduling of the supply air temperature and flow rate to a ventilated slab thermal storage system as well as with a number of numerical test functions. All results demonstrate its superiority compared with the case when CoG is not used