Optimization Design For High Pressure Turbine Casing Typic Structure Based On Thermal-structural Coupling Analysis

In the design of aero-engine casing, the blade tip clearance of high pressure turbine(HPT) was mainly decided by the radial displacement of rotor and casing which was relied on the design of proper structural and cooling air flow. Optimization design was regarded as one of indispensable parts in the aero-engine casing structure design.In this paper, optimization design method for HPT casing axial symmetry model based thermalstructural coupling analysis was studied. Three contents were accomplished as following:(1) The accuracy of thermal-structural coupling analysis method was verified by thermal deformation experiment. In order to obtain the thermal deformation and temperature distribution of casing, a corresponding thermal deformation testing platform was established, and a thermal–structural coupling analysis of casing model in the experimental conditions was performed. The comparision between experimental datas and simulation results showed that: the result got by thermal-structural coupling analysis method used in this paper basically meets the requirement of engineering.(2) An axisymmetric parameterized HPT casing model driven by UG expression file(EXP)was established. Thermal-structural coupling analysis and sensitivity analysis for the model with ANSYS was performed. Based on the sensitivity result, a single-objective and a multi-objective optimization design were proceeded. The optimization results showed that the radial displacement was decreased by 9.85% in the single-objective optimization, and the casing mass was reduced by 4.54% while the radial displacement was decreased by 8.90% in the multi-objective optimization. Obviously, the multi-objective optimization was proved to be better for aero-engine design than the other one.(3) RSM, RBF neural network and kriging model were used to establish casing approximate optimization model. Different setting parameters of each approximate model were studied. And the casing model in the third chapter was optimized with each approximate model. The optimization results showed that the approximate models can get good optimization results, the RBF model is recommended in the single-objective approximate model and the RSM model is recommended in the multi-objective approximate optimization model.