| Swarm intelligence optimization algorithm is an optimization algorithm developed by simulating biological swarm behavior.Because it has no specific requirements to solve the problem and is simple to implement,it has become a widely used method to solve optimization problems in recent years.With the continuous development of swarm intelligence optimization algorithms in recent years,many new swarm intelligence optimization algorithms have emerged.Grey wolf optimization is an emerging swarm intelligence optimization algorithm proposed by Australian scholar Mirjalili in 2014.Mirjalili put forward the grey wolf optimization through mathematical modeling and simulation by observing the living habits and foraging behavior of the grey wolf group.Due to its advantages of fewer parameters and fast convergence,it has become a research hotspot of the majority of Chinese and foreign scholars in recent years.However,because the grey wolf optimization relies on the individual update mode of searching for the optimal solution of the population,when the optimal solution of the population is far from the global optimal solution,the individual keeps approaching the optimal solution of the population and cannot search for the global optimal solution.This results in the stagnation of the search process,leading to premature convergence of the algorithm,and even worse performance in the face of high-dimensional complex problems.Aiming at this shortcoming of the grey wolf optimization,this paper improves the basic grey wolf optimization and applies it to the flexible job shop scheduling problem and the hybrid flow shop scheduling problem these two different types of shop scheduling problems to verify the effectiveness of the improved algorithm.The specific research work is as follows:1.Aiming at the flexible job shop scheduling problem of minimizing the maximum completion time,the grey wolf optimization is easy to fall into local optimum and premature convergence when solving large-scale complex problems,an improved grey wolf optimization is proposed.Firstly,a two-stage coding method based on random key is used to encode the individuals,and an initial population method based on heuristic rules is used to improve the quality of the initial population.Then,in order to balance the process of global search and local search,a nonlinear convergence factor formula based on hyperbolic tangent function is proposed.According to the characteristics of flexible job shop scheduling problem,a weight method based on fitness value is proposed in the individual update stage.Finally,a variable neighborhood search algorithm based on the critical path is embedded in the algorithm decision layer to improve the local search ability of the algorithm.The simulation results show that the improvement of the algorithm can improve the problem that is easy to fall into local optimal to a certain extent,and it is effective in solving flexible job shop scheduling problem.2.In order to minimize the maximum completion time,a discrete random walk gray wolf optimization is proposed for the uncorrelated parallel machine hybrid flow shop scheduling problem.According to the characteristics of the grey wolf optimization,combined with the characteristics of the hybrid flow shop scheduling problem,a cross-based discrete individual update strategy is proposed.In order to increase the probability of finding the global optimal solution,a discrete random walk individual update strategy of grey wolf was proposed.For the inferior solutions that may be produced by the above two methods,the Metropolis criterion of the simulated annealing algorithm is introduced to receive it,so as to increase the ability of the algorithm to jump out of the local optimum.Finally,simulation experiments are carried out based on actual examples and compared with classic algorithms and popular algorithms.The results show that the algorithm can obtain a better scheduling scheme and has the potential to solve the actual hybrid flow shop scheduling problem. |
