Research Of Optimal Design Based On Probability Clonal Selection Particle Swarm Algorithm

The optimal design is a subject developed from the early 60 years of 20th century, which combined the theory of optimization in mathematics and the design of engineering. When people deal with the problems of design, they can find an optimal or a perfect project, which can improve the efficiency and the quality of the design greatly. At present, the optimal design is an important method in the engineering design, which has been used widely in the domain of engineering and has produced enormous benefits in economy and in society. A commingled algorithm--Probability Clonal Selection Particle Swarm Algorithm is put forward. The commingled algorithm is used to carry out the optimal design of the locomotive wheel profile after the analysis of its feasibility. The data of the optimization is simulated and analyzed.First of all, the optimal problem solution technology is studied, and the advantages and disadvantages of the optimal design based on algorithms are analyzed. The successful applications of three intelligent optimal algorithms (immune clonal selection algorithm, particle swarm algorithm, simulation annealing algorithm) are introduced. Then the research content of this thesis is put forward.Secondly, in the domain of optimization, the operational process and characteristics of the three intelligent optimal algorithms are introduced, and their characteristics and shortcomings are analyzed.The single algorithm above has the low solution efficiency and is easy to get the local optimal result. The solution cannot be satisfied.Thirdly, the theory of the commingled algorithm is analyzed, and all of the three algorithms are improved also. On the base of the developments and the characteristics, a new commingled algorithm that is Probability Clonal Selection Particle Swarm Algorithm is put forward, and the application process of the commingled algorithm is presented after the commingled theory is analyzed.Fourthly, the feasibility of the commingled algorithm is analyzed. Based on the theory of the mechanical fuzzy reliability, a cylindrical helix spring is optimized using the commingled algorithm. It is proved that the commingled algorithm is efficient after the comparison among the complex algorithm, genetic algorithm, single particle swarm algorithm and the commingled algorithm. The result shows that the commingled algorithm is feasible.Fifthly, in order to solve the optimal problem of the locomotive wheel profile, the function between the rolling radius difference and the lateral displacement is defined, the function model of profile optimization is made up. Then the optimal design is carried out by means of the commingled algorithm, and an optimal result is achieved.Sixthly, a model of locomotive is assembled in ADAMS/RAIL. Then the simulation is implemented on the experiment line, and the optimal result is gained from the wheel profile data and the commingled algorithm is verified.Finally, a conclusion of the research is drawn and the further study directions are pointed out.