Research And Improvement Of Chimp Optimization Algorithm

As a simple and effective meta-heuristic Algorithm,Chimp Optimization Algorithm(ChOA)has been widely used to solve complex global optimization problems.However,when solving optimization problems,ChOA algorithm also has some shortcomings,such as:precocious convergence,easy to fall into local optimal,can not balance global exploration and local exploitation well.To solve these problems,this thesis has done the following work:(1)In order to improve the ability of ChOA algorithm to solve optimization problems,an Adaptive Chimp optimization algorithm(AChOA)is proposed.In AChOA,the Chimp population is initialized by using Tent chaotic mapping to obtain a better initial solution.At the same time,the adaptive parameters are added in the global search stage,which are changed according to the number of iterations and the population diversity to improve the population diversity.Thirdly,Levy flight strategy is introduced into the position update formula to make up for the shortcomings of ChOA algorithm,such as lack of balance between exploitation and exploration.Finally,to verify the performance of the proposed AChOA,it is compared with 10 well-known algorithms on 19 benchmark functions.Experimental results and statistical analysis show that AChOA has significant advantages over other algorithms.(2)A multi-strategy improved Chimp optimization algorithm(MChOA)is proposed.First of all,the hunting strategy under the state of hunger is designed to expand the search area.Then,adaptive factors are added to the algorithm to enhance the ability of each individual Chimp to actively search and better coordinate global exploration and local exploitation.Finally,the convergence factor and rotation matrix are added in the exploitation stage to improve the ability of the algorithm to jump out of the local optimal.The performance of MChOA is verified by solving 19 reference functions.The experimental results and statistical analysis show that MChOA can effectively improve the precision and convergence speed with the increase of dimension,and the optimization effect is better than 10 comparison algorithms.