Dynamic Probabilistic Design Approach of High-Pressure Turbine Blade-Tip Radial Running Clearance
Cheng-Wei Fei,Yat-Sze Choy,Dian-Yin Hu,Guang-Chen Bai,Wen-Zhong Tang
DOI: https://doi.org/10.1007/s11071-016-2883-1
IF: 5.741
2016-01-01
Nonlinear Dynamics
Abstract:To develop the high performance and high reliability of turbomachinery just like an aeroengine, distributed collaborative time-varying least squares support vector machine (LSSVM) (called as DC-T-LSSVM) method was proposed for the dynamic probabilistic analysis of high-pressure turbine blade-tip radial running clearance (BTRRC). For structural transient probabilistic analysis, time-varying LSSVM (called as T-LSSVM) method was developed by improving LSSVM, and the mathematical model of the T-LSSVM was established. The mathematical model of DC-T-LSSVM was built based on T-LSSVM and distributed collaborative strategy. Through the dynamic probabilistic analysis of BTRRC with respect to the nonlinearity of material property and the dynamics of thermal load and centrifugal force load, the probabilistic distributions and features of different influential parameters on BTRRC, such as rotational speed, the temperature of gas, expansion coefficients, the surface coefficients of heat transfer and the deformations of disk, blade and casing, are obtained. The deformations of turbine disk, blade and casing, the rotational speed and the temperature of gas significantly influence BTRRC. Turbine disk and blade perform the positive effects on the BTRRC, while turbine casing has the negative impact. The comparison of four methods (Monte Carlo method, T-LSSVM, DCERSM and DC-T-LSSVM) reveals that the DC-T-LSSVM reshapes the possibility of the probabilistic analysis of complex turbomachinery and improves the computational efficiency while preserving the accuracy. The efforts offer a useful insight for rapidly designing and optimizing the BTRRC dynamically from a probabilistic perspective.