Research Of The Multi-objective Optimization Problem Based On Immune Genetic Algorithm

In the engineering practice and scientific research, all kinds of multi-objectiveoptimization problems often occur. There are many kinds of methods to solve themulti-objective optimization problems, and the genetic algorithm is one of the typicalmethods. But when using the genetic algorithm to solve the multi-objective optimizationproblem, it is easily trapped into local optimum and can’t keep the diversity of populationvery well. In solving multi-objective optimization problems, antibody concentration in theartificial immune algorithm is used to represent the diversity of solution groups, and memorycells can prevent the algorithm into premature convergence effectively.In order to solve the multi-objective problem, the immune thought is applied to the non-dominated sorting genetic algorithm. A non-dominated sorting genetic algorithm based onimmune theory is put forward. The memory cells on the basis of genetic algorithm areredefined, and a new kind of concentration calculation method and the adaptive mutationoperator are designed. Memory cells consist of non dominated sorting sequence number and asmaller number of antibodies, which is used for the good solutions and related parameters ofthe retention issues. New method to calculate the concentration can effectively prevent theloss of good solutions. The designed mutation operator can make antibodies to adaptivelyadjust the mutation probability based on the concentration, which prevents the algorithm intopremature convergence.The proposed algorithm is tested with four different multi-objective test functions, thetest results show that it can effectively make Pareto solution evenly distributed. The structuralparameter optimization of the tooth-arrangement multi-fingered dextrous hand is amulti-objective optimization problem. The proposed method is applied into this problem. Theresults show that it can effectively solve the optimization problem. The non-dominated sorting genetic algorithm based on immune principle can make full use of good ideas in the geneticalgorithm and immune algorithm, which has great significance for the optimization theoryand the real life.