Improved Grey Wolf Optimization Algorithm And Its Application Research

The rapid development of society is benefited from the promotion of various cuttingedge technologies,and the power of technology has deeply influenced and changed people’s daily lives.As time goes by,solving various optimization problems has become increasingly difficult,making traditional algorithms unable to meet the needs of realworld and potential requirements.In recent years,researchers have found that heuristic algorithms have unique advantages in solving optimization problems,providing us with a new way of thinking and direction.Grey wolf optimization algorithm is a heuristic algorithm with a clear structure,few configuration parameters,and easy implementation.However,this algorithm still has problems such as low convergence accuracy and easy entrapment in local optimal solutions.Therefore,it is necessary to make appropriate improvements and optimizations.Based on the grey wolf algorithm,this study aims to optimize its performance and propose new improved algorithms to solve engineering design optimization problems and path planning problems.The main research work of this paper includes the following aspects:(1)An improved grey wolf optimization algorithm(SCGWO)based on chaotic mapping,sine cosine search strategy,and chaotic perturbation is proposed.The first step is to initialize the grey wolf population with chaotic mapping to enhance the uniformity and exploration of the population,thereby improving the optimization accuracy.Next,the sine cosine search strategy is introduced in the subsequent search process to perform secondary optimization on the optimal result,thereby breaking through local optima and enhancing the global search ability of the grey wolf algorithm,thus improving the overall performance of the algorithm.Finally,the chaotic perturbation strategy is introduced to avoid premature convergence and local optimal solution and to enhance the global exploration and exploitation performance of the algorithm.The improved algorithm and four representative algorithms were tested on 18 benchmark functions,and the experimental results showed that the SCGWO algorithm had better solving ability than the other four compared algorithms.(2)A Grey Wolf Optimization algorithm based on oppositional search strategy and Gaussian mutation under Levy flight(LGGWO)is proposed.Firstly,to address the convergence speed issue of the basic Grey Wolf Optimization algorithm,an oppositional search strategy is introduced when initializing the Grey Wolf population.To improve the accuracy of the algorithm,it is necessary to help balance the global and local search phases.When the algorithm is trapped in a local optimal solution,the introduction of the Levy flight strategy can help the algorithm quickly jump out and further improve its performance.Finally,the Gaussian mutation strategy is adopted to maintain the diversity of the population during the evolution process of the algorithm,thereby increasing the stability and adaptability of the algorithm.Using a greedy selection approach to update the population,successfully solved the problem of the algorithm’s difficulty in escaping local optimal solutions in the later stages.The improved algorithm is compared with three other representative algorithms and tested on 23 benchmark functions.The experimental results show that the LGGWO algorithm outperforms other comparison algorithms significantly in both solving ability and application potential.(3)To further validate the authenticity and practicality of the two proposed algorithms mentioned above,the improved SCGWO algorithm and LGGWO algorithm are respectively applied to the research of engineering design constraint optimization problems and path optimization problems.Through simulation and comparative experiments,it can be observed that the proposed algorithms not only have significant optimization performance but also have good application effects.