Multi-strategy Integrated Harris Hawk Algorithm to Solve Global Optimization Problems
In basic science and practical engineering applications,there are problems of solving optimization schemes in different dimensions and under multiple constraints,and it is difficult for most conventional methods to deal with such optimization problems effectively.Intelligent optimization algorithms are able to solve many optimization problems in the case of failure of classical optimization techniques due to their small search space,few search times,flexible computation and strong applicability.With the widespread application of intelligent optimization algorithms in many fields such as logistics scheduling,combinatorial optimization,system control,etc.,more and more scholars have begun to study such algorithms.Most of these algorithms are designed by the influence of biological and physical phenomena in nature,and are increasingly used in the engineering field due to their advantages of simple concept and easy implementation.Harris Hawk Optimization(HHO)was proposed in 2019 as a new type of intelligent optimization algorithm,which was inspired by the hunting behavior of the Harris Hawk,and has the advantages of simple principle,easy programming,fewer parameters,high conver-gence accuracy and fast convergence,and has been applied to the design and engineering optimization problems in certain disciplines.For different types of function optimization problems and engineering applications,the HHO algorithm has the problems of slow convergence speed and insufficient stability of optimization search.In order to further improve the performance of the Harris Hawk algorithm in solving problems,this paper proposes a multi-strategy Improved Harris Hawk Optimization(IHHO)algorithm that integrates the good point set,nonlinear energy escape factor,and Logistic-Cubic cascading chaotic perturbations.Firstly,for the characteristics of random generation of the initial population,the good point set strategy is applied for optimization to uniformly distribute the initial population and improve its traversal ability.Secondly,for the problem that the algorithm is easy to fall into local optimal solution,a nonlinear energy escape factor is proposed based on the different characteristics of each stage of the algorithm,and the escape factor changes from large to small according to the number of iterations,i.e.,expanding the search range in the early iteration to prevent the algorithm from falling into local optimal,and reducing the search range in the late iteration to accelerate the convergence of the algorithm,so as to balance the algorithm's global and local exploration ability.Finally,for the problem that the search position is easy to converge locally,Logistic-Cubic cascade chaos is introduced to perturb the search position during the updating process,to avoid the algorithm from falling into local optimum,and to improve the solution accuracy and convergence speed.In the simulation experiment stage,the IHHO algorithm is used to solve 23 function problems with different characteristics,each problem is solved 30 times,and the mean and standard deviation of each result is taken to compare with the other 7 algorithms.And the results are verified by using the target convergence curve and Wilcoxon rank sum test.The results indicate that the IHHO algorithm has stronger optimization performance and solution stability than other algorithms.At the same time,the IHHO algorithm is used to optimize the solution of the three-truss design engineering problem,and the results show that the algorithm has strong competitiveness com-pared to the comparative algorithms,and has the ability to become an effective tool for solving global optimization problems.In the future,further research will be conducted on intelligent optimization algorithms,combining them with deep reinforcement learning to further solve larger and more complex practical application problems.
Harris Hawk optimization algorithmgood point set strategynonlinear escape factorcascade chaosengineering problems
万方数据
维普
