FGA-FEM Method And Structure Optimization Of Floatwall In Combustor

The floatwall in combustor is an important structure for aero-engine to improve its power-to-weight. FGA-FEM was put forward to optimize the floatwall structure, and FGA-FEM combined Fuzzy Genetic Algorithm (FGA) and Finite Element Method (FEM). The study work mainly included the following:(1) To overcome the premature convergence of the Standard Genetic Algorithm (SGA), a genetic algorithm (GA) based on fuzzy reasoning was proposed. This method was used to adjust the probabilities of crossover and mutation during the evolution of GA. The results of optimization on mathematical functions showed that the shortcoming of SGA can be overcome, and the relying on the initialization of parameters can be effectively avoided. The quality of the solution and the convergent speed also can be improved.(2) FGA-FEM was used to optimize the poles'positions on the flat structure, which was a simple model of the floatwall structure. The results show that the method of FGA-FEM has the better performance.(3) A tactic of the optimization design was adopted, which was achieved by adding the variable and the complex conditions step by step. First, the optimization of the poles'positions will be obtained. Second, the optimization of the poles'area based on the results of the first step can be reached. Finally, the optimization of the thickness of the flat based on the results above will finish the whole optimization process. The method and the tactic are proved to be efficiency by the result of the floatwall structure's optimization.(4) Based on ANSYS software, the code of FGA-FEM method was written in VC++. The computing software system can be transplanted easily. It has the excellent analysis ability, and so on.