Improvement Of Artificial Immune Algorithm And Its Application Research

Artificial immune algorithm is a new intelligent optimization algorithm based on immune evolution mechanism and information processing mechanism of biological immune system.The algorithm has many excellent characteristics of the biological immune system and the relevant research results are widely applied in many fields.Therefore,how to make use of these excellent features to improve the performance of other intelligent algorithms and solve the new problems in engineering practice is worth studying.In this thesis,a new immune particle swarm optimization algorithm is proposed based on the in-depth analysis and research of the principle,mechanism and performance of biological immunity and the artificial immune algorithm.The main research works of this thesis as following:(1)To give the general framework of immune algorithm and research the immune principle of various immune algorithms deeply.According to specific characteristics of the problem,a new method for the selection of antibody concentration and the operation of immune vaccine is propose to improve the efficiency of the algorithm.(2)An improved immune particle swarm optimization algorithm is proposed(IPSO).The new algorithm can effectively overcome the inherent shortcomings of the particle swarm algorithm by introducing antibody concentration regulation mechanism,diversity maintenance mechanism and immune selection operation.The feasibility and effectiveness of the improved algorithm are proved by experiments.(3)The IPSO algorithm is applied to the flexible job shop scheduling problem(FJSP)and a mathematical model of FJSP is given.According to the characteristics of FJSP model,a new method of antibody coding,antibody update operator,antibody selection mechanism and the selection of vaccine and vaccination methods were designed.Finally,the effectiveness and superiority of the algorithm are verified by several experiments.