Research And Application Of Multi-strategy Improved Harris Hawk Algorithm

Harris Hawk Algorithm(HHO)is a new bionic swarm intelligence optimization algorithm proposed in recent years.Its algorithm simulates the process of finding the global optimal solution in the process of prey hunting by a flock of hawks.The algorithm has the advantages of high plasticity,strong local finding ability,and good performance in finding the optimal solution,and is widely used in the fields of multivariate function optimization,nonlinear programming,and optimal design.However,the Harris Hawk algorithm suffers from a series of problems such as insufficient exploitation ability,decreasing population diversity,and easily falling into local optimality,etc.In order to better improve the theoretical basis and application of Harris Hawk algorithm,this thesis conducts a research on multistrategy improvement of Harris Hawk algorithm and validates it effectively in practical problems.The main work of this paper is as follows:(1)A chaotic Harris Hawk algorithm(AMHHO)based on whale hunting mechanism is proposed.The improved Tent chaotic mapping is used to improve the initialization phase of the Harris Hawk population to increase the population diversity;the perturbation of the updated position is combined with the dynamic spiral bubble net hunting method to improve the search accuracy of the algorithm;the adaptive t-distribution and Gaussian variance perturbation are introduced to strengthen the algorithm’s ability to jump out of the local optimum.The benchmark function test shows that the improved algorithm has greater improvement in benchmark function search than other comparative algorithms.(2)An elite Harris Hawk algorithm(m-HHO)combining distributed estimation strategies is proposed.A chaotic local search strategy is introduced in HHO to exploit the advantages of chaotic mapping to improve the exploitation capability of the algorithm;an elite alternative pooling strategy is proposed to enhance population diversity;a distributed estimation strategy is used to modify the evolutionary direction of the algorithm to improve the convergence efficiency of the algorithm.CEC2017 test experiments show that the improved algorithm takes into account the convergence speed and global search and other capabilities,and finally the algorithm is used to solve engineering Finally,the algorithm is used to solve engineering constraint problems to demonstrate the practicality of the optimized algorithm.(3)A two-stage backward learning strategy optimization-based HHO algorithm(TSOHHO)is proposed.A two-stage reverse learning strategy search strategy is introduced to balance and enhance the development and exploration ability of the algorithm and accelerate the convergence of the algorithm;a jump strength update strategy combined with nonlinear escape energy decreasing control is proposed to effectively realize the transition from global exploration to local development of the algorithm;a stochastic learning strategy is added to enhance the communication among individuals,thus avoiding the algorithm from falling into local optimum.The performance of TSOHHO is verified by the CEC2017 test set and the ability of the algorithm to solve practical optimization problems is tested in a robot path planning problem.