Research On The Improvement And Application Of Slime Mold Algorith

Optimization problems widely exist in many fields such as artificial intelligence,management decision-making,and engineering design.With the development of science and technology and the continuous expansion of the social production scale,optimization problems have become more and more common and complex.Traditional derivative-based exact algorithms are not suitable for solving non-derivable,discontinuous,large-scale,nonlinear,and dynamically changing optimization problems.Metaheuristics are the first choice for solving such complex problems.The slime mould algorithm(SMA)is a newly proposed meta-heuristic algorithm based on swarm intelligence.It has the advantages of simple and flexible algorithm structure and outstanding optimization performance,compared with other intelligent algorithms.However,it also has shortcomings.For example,when faced with some complex problems,the convergence speed of SMA will slow down,and it will easily fall into a local optimal solution and the imbalance between global exploitation and local exploration.This paper integrates three improved strategies into the SMA to improve the diversity of its population,the convergence speed,the quality of the final solution,the ability to jump out of the local optimum,and the robustness performance of basic SMA.The proposed improved SMA is applied to solve three types of real-world engineering optimization application problems,and good application effects have been achieved.The main research contributions of this paper include:(1)Based on the basic SMA,an orthogonal learning strategy is introduced to enable individuals of slime mould to have orthogonal learning capabilities,improve the convergence rate of the algorithm,and the quality of the final solution.At the same time,a chaotic initialization strategy is introduced to increase the diversity of the initial population of the algorithm to enhance the population richness and optimization ability.In addition,a new boundary reset strategy is introduced to further enhance the population diversity and global search ability of the algorithm,as well as the ability to jump out of the local optimal solution.(2)The proposed improved SMA is applied to solve the IEEE CEC2014 benchmark function problem,and compared with several other advanced metaheuristics.The simulation results show that the improved SMA has better optimization results,and the overall optimization performance is better than the comparison algorithms.The convergence speed,the quality of the final solution,and the structure of the solution are significantly improved over the basic SMA.(3)Finally,the proposed improved SMA is applied to two types of real-world engineering optimization problems.The first is to solve the adaptive IIR filter model identification problem.Compared with some other advanced meta-heuristic algorithms,the algorithm proposed in this paper has better simulation results.To further broaden the application research of the algorithm,the proposed algorithm is applied to solve the welded beam design problem and the pressure vessel design problem.Compared with some well-known algorithms,the proposed improved slime mold algorithm achieves better optimization results and shows higher optimization performance on these two design problems.The excellent application in three application problems proves that the proposed improved SMA in this paper has excellent performance and the effectiveness and practicability of solving real-world engineering optimization problems.