A Hybrid Grasshopper Optimization Algorithm Based on Simulated Annealing

Yang Cong,Huan Li,Wenhong Wei

DOI: https://doi.org/10.18178/wcse.2021.06.024

2021-01-01

Abstract:This paper proposed a hybrid grasshopper optimization algorithm to overcome the disadvantages that the grasshopper optimization algorithm was easy to fall into local optimal solution and low accuracy.Firstly, this work used reverse learning strategy to generate the initial population to enhance the global search efficiency and the quality of the solution; secondly, the dynamic compression factor is introduced to replace the linear adaptation of the key parameters in the basic grasshopper optimization algorithm to enhance the global search ability of the algorithm; finally, this paper adapts to the metropolis receiving criterion of simulated annealing algorithm to receive the poor solution with a certain probability, so that the algorithm can be used Enough to jump out of the local optimal solution.Experiments show that the hybrid grasshopper optimization algorithm has stronger global search ability, better accuracy, and can effectively jump out of the local optimal solution.