A New Fuzzy Adaptive Simulated Annealing Genetic Algorithm and Its Convergence Analysis and Convergence Rate Estimation

摘要

Due to shortcomings of genetic algorithm that its convergence speed is slow and it is often premature convergence, a new improved genetic algorithm - fuzzy adaptive simulated annealing genetic algorithm (FASAGA) is presented by integrating fuzzy inference, simulated annealing algorithm and adaptive mechanism. The strong Markovian property attributed to the population sequence was deduced by mathematical modeling. Then the convergence in probability of the FASAGA was proved on the condition that the time tended to infinity. Then convergence speed of FASAGA was estimated and some quantitative results were achieved. The simulation results validated the theoretical analysis conclusions. This work is helpful to further analyze and improve optimization performance of FASA-GA and other hybrid genetic algorithms.