Study On The Improved Cultural Algorithm And Its Application

Science and technology develop fast and the study on all kinds of science fieldsadvance by leaps and bounds，and all kinds of optimization problems met in the actualmanufacture and life are more and more complex, the methodology that is used toaddress the problems needs to be improved greatly. The single intelligent optimizationalgorithm is used widely, such as genetic algorithm, cultural algorithm, particle swarmoptimization algorithm, immune algorithm etc. But for the high standard requirement ofengineer application, the disadvantage of these algorithms is more remarkable. Thusmany improved algorithms emerge as the right moment, such as improved geneticalgorithm, improved cultural algorithm, the cultural genetic algorithm, the particleswarm genetic algorithm, the immune genetic algorithm, the particle swarmoptimization algorithm based on the cultural algorithm etc. If combining two algorithmsingeniously, the improved algorithm owned the advantages of two algorithms enable tomake up for the shortcoming of each algorithm, and then it will meet the demands ofactual application. The effective and reasonable target assignment plays a crucial role inthe complex air defense system, while optimization algorithm is very efficient methodachieving the target optimum assignment. Presently, many experts focus on studyingthis field. To guide missile and attack targets, array antenna is of great importance. Andintelligent optimization algorithm plays a crucial role in antenna optimization toachieving specific function. Thus this paper does some improvement about the currentoptimization algorithm, and applies the improved algorithm to address the problem ofthe missile target allocation and synthesis of radar array antenna. Finally somefollow-up improvement is achieved in the novel algorithm above.Focusing on addressing the existing problems of genetic algorithm in the missiletarget allocation and the optimization of radar array antenna, this paper analyzesdisadvantage of intelligent algorithm in optimizing these problem above, and thencultural algorithm is improved again and again, finally, the improved algorithm areapplied to practical optimization problem to achieve better effort. Aiming to theproblem of missile target assignment, cultural algorithm in which genetic algorithm isused to be its population space, besides combined with truncation selection and reverseoperation is used to optimize classical function and the model of target allocation.Simulation results show that the improved algorithm is more effective than that improved before, besides it has a great advantage in convergence rate, evolutiongeneration and convergence speed. Hence the novel algorithm is an effective algorithmin addressing the problem of air defense missile optimum target assignment. Accordingto the optimization problem of radar array antenna, this paper introduces better acceptfunction, selection function and update function to improved cultural algorithm which iscombined with multi-population mechanism. The results illustrate that the newalgorithm has strong capability of global search and local search. Finally the best phaseallocation is achieved and the side lobe level is reduced effectively. The studycontributes great effect to the optimization of array antenna. In order to further improvethe performance of the above algorithm, this paper improved the algorithm formpopulation interior and between populations at last. Adaptive crossover probability andmutation probability is used within each population, meanwhile competitive mechanismis used between populations. The two mechanisms improve population diversityeffectively, and balance local search ability and global search ability. The result showsthat the novel algorithm has a faster speed of convergence and higher rate ofconvergence compared with improved before.