Research On The Methods Of Tolerance Optimal Design For Aircraft Assembly

In the process of design and manufacture of aircraft, it is very important that the good tolerance can improve the assembly performance, decrease manufacturing cost, and improve the success rate of assembly. This paper is based on the current status of tolerance design of aircraft parts. The methods of tolerance optimal design for aircraft assembly are researched in this paper. The key research works are as follows.Tolerance analysis process is established and Monte-Carlo Simulation is mainly studied in this paper for tolerance analysis. Simulation results indicate that the result of tolerance analysis with Monte Carlo Method is close to the statistical tolerancing methods and the calculation precision is improved as the simulation is increased. However, Monte-Carlo Simulation does not require a lot of statistical data. In addition, when the size distribution does not obey normal distribution, it also can be calculated by Monte-Carlo Simulation. When the success rate of assembly is too poor, the methods for adjusting divided link are studied. Lastly, it is illustrated the assembly tolerance analysis process through an example.Tolerance optimal allocation process is established and the mathematical model of optimizing assembly tolerance is studied in this paper. Based on real-coded genetic algorithm, this thesis researches into optimal allocation of tolerances because tolerance allocation model for the optimization is complex non-linear characteristics. A modified particle swarm optimization algorithm is proposed to improve the performance of real-coded genetic algorithm. A modified dynamic particle swarm optimization algorithm that is applied to optimal allocation of tolerances is researched. It is illustrated by an instance that the solving efficiency of modified particle swarm optimization algorithm is superior to particle swarm optimization algorithm and real-coded genetic algorithm. The convergence,stability and efficiency of the modified particle swarm optimization algorithm is markedly improved.Having established tolerance optimal allocation process, the prototype system of computer aided aircraft tolerance design (CAAT) was given out which is developed on the theories studied before with Matlab GUI and Access. Specific contents of the prototype system are tolerance Automatic Query Module, tolerance analysis module and tolerance allocation module. The prototype system is verified by the instance.