Improved Swarm Intelligence Optimization Algorithm And Its Application

With the rapid development of science and technology,optimization problems are involved in many fields,such as travel salesman problem,functional optimization,power dispatching,truss structure design,etc.However,many optimization problems are multimodal,high-dimensional,and difficult to solve.Traditional optimization algorithms such as conjugate gradient method,simplex method,Newton method have been difficult to meet people's needs.Therefore,the design of efficient optimization algorithm has become the research goal of many researchers.This paper mainly studies swarm intelligence optimization algorithm,improves the advantages and disadvantages of each optimization algorithm,and applies it to practical problems,so as to make it play its application value.This paper mainly improves the artificial bee colony algorithm and applies it to the practical field.The specific research contents of this paper are as follows:First,comparing various popular algorithms to provide theoretical support for the subsequent improved artificial bee colony algorithm.Secondly,aiming at the characteristics of strong global search ability and weak local search ability of traditional artificial bee swarm algorithm,this paper proposes a new strategy to retain the historical optimal solution group of the whole population search.On this basis,the search formula of the employing bee stage and onlooker bee stage of the artificial bee colony algorithm is improved,and the performance of the improved artificial bee colony algorithm(MNABC)is tested experimentally on the standard test function and the practical problem of the economic power dispatching.The experimental results show that the proposed algorithm is efficient and reliable.Thirdly,aiming at the characteristics of slow convergence speed,low accuracy in the later stage,and difficulty in getting into the local optimum when dealing with complex optimization problems,inspired by particle swarm optimization,this paper provides an intelligent learning strategy for the employing bee stage and onlooker bee stage of the artificial bee colony algorithm.In addition,according to the characteristics of scouts during the scout phase.This paper propses a novel learning strategy.Then,the improved algorithm(EABC)is tested on the standard test function and the practical problem of truss structure design for the experimental test of its performance,and the reliability of the algorithm is verified by comparing with the current popular algorithms.In this paper,in view of the slow convergence speed of the artificial bee colony algorithm and its weak local search ability in the late iteration,two improved algorithms are proposed,namely artificial bee colony algorithm based on the intelligent strategies(MNABC)and artificial bee colony algorithm based on the elite search strategies(EABC),then these two algorithms are applied to practical problems,in which the MNABC algorithm is applied to the power economic dispatch problem,the EABC algorithm is applied to the truss structure design problem,the experimental results verify the effectiveness of the algorithm.