Research On Optimization And Application Of Tunicate Swarm Algorithm

Tunicate Swarm Algorithm(TSA)is an emerging swarm intelligence optimization algorithm inspired by the migration process of deep-sea ensembles,mainly simulating the jet propulsion and population behavior of ensembles during foraging.The tunicate swarm algorithm has the advantages of simple structure,fewer parameters,and strong search ability.However,the research on the tunicate swarm algorithm is still in its preliminary stage,and there are some shortcomings in the algorithm itself,such as low solving accuracy,susceptibility to falling into local optima,and slow convergence speed.This thesis aims to conduct an in-depth analysis of the shortcomings of the tunicate swarm algorithm and propose two improved tunicate swarm algorithms.At the same time,the improved algorithm will be applied to practical fields.The main research work of the thesis are as follows:(1)Aimed at the problem that the tunicate swarm algorithm is easy to fall into local extremum,the improved tunicate swarm algorithm(ELTSA)based on opposition based learning and levy flight is proposed to improve the solution accuracy and optimization ability of the algorithm.First,benchmark test functions were used to test and verify the improved tunicate swarm algorithm’s strong solving ability.Second,the improved tunicate swarm algorithm was further utilized to optimize the deployment of wireless sensor nodes,and the results showed that the improved tunicate swarm algorithm has significant advantages in solving the wireless sensor node deployment problem.(2)Propose an integrated logistic chaotic mapping and quadratic interpolation-based tunicate swarm algorithm(ITSA)to balance the development and exploration abilities of the TSA algorithm.Apply ITSA algorithm to solve the image segmentation problem of K-means,and through experimental simulations,compare the test results with four other algorithms to further demonstrate that ITSA algorithm has better optimization ability in solving such problems.(3)It is proposed to apply the ITSA algorithm to five constrained optimization problems,including wheel system design,rolling bearing design,and cantilever beam design.The results show that the ITSA algorithm can achieve good competitiveness in engineering design optimization problems,and can therefore be used as an alternative optimizer for solving optimization problems in current intelligent optimization algorithms.In view of the weak local search ability of the tunicate swarm algorithm,this thesis proposes two improved tunicate swarm algorithms,namely,the tunicate swarm algorithm based on opposition based learning and levy flight(ELTSA)and the tunicate swarm algorithm integrating logistic chaotic mapping and quadratic interpolation(ITSA).And these two algorithms were applied to practical problems,and the experimental results verified the effectiveness of the algorithm,which has certain theoretical value and practical significance.