Study On Multi-objective Optimization Of Engineering Project Based On Immune Genetic Algorithm

With the development of technique of engineering management andmulti-objective optimization (MO), more and more multi-objective technique is usedon the management on engineering project. There are lots of defects about classicalgorithms which have been used in MO field when dealing with complicatedproblems. Nowadays with the development and application of intelligence algorithmin science and experiment, the application of intelligence algorithm is a hot researchpoint in dealing with MO problems.Based on the aboard studying on the in and abroad related literatures, deep theoryresearch and test analyses have been done on multi-objective genetic and AIA. Afteranalyze the characters of both algorithms, this paper brings out a MO algorithm basedon immune and genetic in dealing with MO problems about engineering .The contentof the research is as follows:The thesis summarizes the existing multi-objective genetic algorithm, analyzingthe flow, the strategy of assigning fitness, the strongpoint and weakness of eachalgorithm. Based on the research of the theory of AIA, this paper also brings out theflow of the average AIA, and then does some improvement. After comparing the AIAbased on between the information entropy and Europe distance, analyzing thecalculate speed and the converge efficiencyof the two algorithms.Besides, the paper brings out the average model of immune genetic algorithm,after combining the theory of this two algorithms, and analyzing the strategy ofassigning fitness, the paper brings out improved immune genetic algorithm, whichapplies the assignment fitness strategy of NSGA, combining the small environmenttechnique and concentration mechanic of AIA, so that it could retain the diversity ofthe antibody group. The method calculating the concentration based on the Europedistance is better than that based on information entropy in the celculate speed,furthermore, the memory cell mechanic guarantees that most of the Pareto solutionscould be found out. The efficiency of this algorithm is approved by the multi-apexfunction.To testifythe availability of the algorithm, a engineering project multi-objectivemodel is introduced, the improved algorithm is applied in dealing with the optimization of a reality problem, then analyzing the outcome, which approves theefficiencyof this algorithm.