Improved adaptive genetic simulated annealing algorithm

Abstract

This paper aims to use an improved version of the real coding genetic algorithm to achieve a globally optimal solution to complex problems, such as the Traveling Salesman Problem (TSP). The Genetic Algorithm (GA) is prone to fall into local optimal solutions, and the Simulated Annealing algorithm (SA) converges slowly. In this paper, an optimization algorithm based on an improved Adaptive Genetic Simulated Annealing Algorithm (AGSAA) is proposed. Then an adaptive crossover and mutation probability is improved, which can effectively avoid the algorithm from falling into local optimum. Finally, a simulated annealing operator is added according to the evolutionary process of the algorithm, and an adaptive Metropolis criterion and a minimum temperature are improved to make the algorithm more adaptive. The experimental results on the TSP example show that the proposed AGSAA can obtain better optimization results compared with the results of other optimization algorithms.