Research On Multi-objective Optimization AlgorithmBased On Chaotic Birds Swarm Algorithm

In practical applications and scientific researches,it is needed to optimize certain issues accompanied by constraints with some purposes regularly.And simultaneously these optimization problems also have a plurality of targets which need to be optimized.Therefore,the mathematical model with multiple objective functions replaces this model of a single target which is not sufficient to describe all the characteristics of these problems.Investigating the solving method to the optimization problems with multiple objective functions has academic significance and application value.It is called multi-objective intelligent optimization when applying intelligent optimization method to solve the optimization with multi-objective function.Due to its advantages,it is widely used in engineering and a wide range of disciplines.Based on the new Birds Swarm Algorithm,this paper focuses on the multi-objective evolutionary algorithm,and its highlights include the following two points:Because the original bird swarm algorithm is easy to fall into local optimization and poor stability when dealing with high dimensional variable optimization problems.,in this paper,a bird swarm optimization algorithm(CBSA)based on chaos is proposed,which can improve the original bird swarm optimization algorithm.In the initialization,CBSA selected individuals in the whole solution space of the optimization problem by logical mapping,and and after a certain number of iterations,it added a small amplitude chaotic disturbance on the optimal solution to avoid the algorithm into a local optimumz.Through seven classic benchmark functions,the algorithm and the classical PSO,BA and initial birds algorithm are compared at the low-dimensional and high dimension in order to verify the superiority and stability of CBSA.In addition,a multi-objective chaotic bird swarm optimization algorithm(MOCB SA)is proposed when the chaotic bird swarm algorithm is successfully applied to the multi-objective optimization problem.MOCBSA used the search mechanism of the chaotic bird swarm algorithm,maintained the non-dominated solution set by the external elite set,and maintained the distribution of the Pareto optimal solution set by the comprehensive sorting operation and improved crowding distance.Through simulation experiment on low-dimensional,high-dimensional ZDT series test function,and DTLZ series of test functions,compared MOCBSA with classic NSGA-? algorithm,MOEA/D algorithm and MOPSO algorithm,then evaluated the performance metrics GD,SP,C and the convergence curve,it proves that the algorithm is feasible.