An improved RNA genetic algorithm without mutation operation in the later stage

In the late stage of the population evolution of the RNA genetic algorithm, mutation may destroy the individual genes with higher fitness, resulting in slower convergence or no optimal solution. Therefore, this paper proposes an RNA genetic algorithm that does not add mutation operations in the late stage of evolution (Referred to as nmlsRNA-GA), in the early stage of evolution, the goal is to find the approximate location of the optimal solution. The two-point crossover and fixed-probability mutation operations are performed in the crossover pool composed of individuals with high fitness and individuals with low fitness to increase the diversity of the population. In order to improve the global search ability of the algorithm. In the later stage of the evolution, the goal is to improve the local search ability to determine the optimal solution, and use individuals with high adaptability to cross over individuals with high adaptability to generate individuals with higher adaptability, thereby improving search Accuracy, no mutation operation is added at this stage to prevent mutation from destroying excellent genes. By optimizing five typical high-dimensional test functions, and comparing them with the RNA genetic algorithm that uses mutation throughout the process, the analysis results show that the algorithm has better performance.