An Improved Gray Wolf Optimization Algorithm Based on Levy Flight and Adaptive Strategies

Abstract

A new improved gray wolf optimization algorithm (LAGWO) is proposed to address the disadvantages of global exploration and local exploitation imbalance, slow convergence speed, low optimization-seeking accuracy and easy to fall into local optimality when solving complex problems. Firstly, the influence of the attenuation factor on the gray wolf optimization algorithm is analyzed, and an adaptive attenuation factor with different exploration ratios can be set according to different optimization problems is proposed to balance the exploration and exploitation capabilities of the algorithm and to ensure that the algorithm has a certain global search capability even at the late stage of the optimization search. Numerical simulation experiments show that increasing the exploration capacity ratio is beneficial to improving the convergence accuracy of the algorithm. Then, the characteristics of occasional long-distance walking of Levy's flight are applied to the optimization search process of α and β wolves to improve the global search ability of the algorithm and avoid falling into local optimum. Aiming at the feature that the candidate wolves ignore the different importance of the three leading wolves in the position update, the adaptive learning weight strategy is proposed to ensure that the constraint of individual gray wolves is reduced at the early stage of the algorithm seeking and improve the global search ability of the algorithm, and at the same time, it can speed up the convergence speed and improve the convergence accuracy at the late stage of the seeking. Finally, simulation experiments are carried out for 12 standard test functions and compared with several other algorithms, and the experimental results show that the algorithm has greater advantages in the optimization-seeking accuracy, algorithm stability and convergence speed.