A novel real-coded GA is proposed which utilizes chaotic search and variable evolutionary rate. By introducing these techniques, the GA algorithm can better simulate the process of biologic evolution, and possesses better hill-climbing ability. In addition, "family competition" is added into the process of mutation and the operating order of mutation and crossover operators is adaptively changed in a different evolutionary stage. Compared with self-adaptive GAs, the real-coded GA can overcome the shortcoming of premature convergence and stagnation, and effectively solves the problem of global convergence.
展开▼
