Evolutionary Algorithm for Solving Multi-Objective Optimization Problem Based on Projection Plane with Objective Decomposition
In order to solve the question that the multi-objective optimization problem with many solu-tions,this paper studies how to obtain the required solutions under decision condition,rather than all the solutions of the multi-objective problem.At present,the existing multi-objective algorithms mainly fo-cus on the distribution and convergence of the solutions,and rarely consider the number of solutions.Ac-cording to decision condition,using the multi-objective optimization algorithm based on plane to decom-pose the objective function of the multi-objective optimization problem,and different objective functions are divide into the plane and free dimension,so the objective space under the user's decision conditions is established.In the objective space,limit the objective range of the objective function that constitutes the projection plane in the objective space,use the spatial distance to obtain the fitness of the projection plane,and use the aggregation function in the MOEA/D algorithm based on the decomposition strategy as the fit-ness function on the free dimension.Through related experimental tests and analysis,it is proved that Evo-lutionary Algorithm for solving Multi-objective Optimization Problem Based on Projection Plane with objective decomposition can effectively solve the multi-objective optimization problem of determining the objective domain.
NETL
NSTL
万方数据
