Distributed Collaborative Response Surface Method for Mechanical Dynamic Assembly Reliability Design

Because of the randomness of many impact factors influencing the dynamic assembly relationship of complex machinery, the reliability analysis of dynamic assembly relationship needs to be accomplished considering the randomness from a probabilistic perspective. To improve the accuracy and efficiency of dynamic assembly relationship reliability analysis, the mechanical dynamic assembly reliability（MDAR） theory and a distributed collaborative response surface method（DCRSM） are proposed. The mathematic model of DCRSM is established based on the quadratic response surface function, and verified by the assembly relationship reliability analysis of aeroengine high pressure turbine（HPT） blade-tip radial running clearance（BTRRC）. Through the comparison of the DCRSM, traditional response surface method（RSM） and Monte Carlo Method（MCM）, the results show that the DCRSM is not able to accomplish the computational task which is impossible for the other methods when the number of simulation is more than 100 000 times, but also the computational precision for the DCRSM is basically consistent with the MCM and improved by 0.40-4.63% to the RSM, furthermore, the computational efficiency of DCRSM is up to about 188 times of the MCM and 55 times of the RSM under 10000 times simulations. The DCRSM is demonstrated to be a feasible and effective approach for markedly improving the computational efficiency and accuracy of MDAR analysis. Thus, the proposed research provides the promising theory and method for the MDAR design and optimization, and opens a novel research direction of probabilistic analysis for developing the high-performance and high-reliability of aeroengine.

On the hydrodynamics of hydraulic machinery and flow control

Hydraulic machinery mainly includes turbine and pump, which is closely related to national economy and people＇s livelihood involving aerospace industry, marine engineering, hydropower engineering, petroleum industry, chemical industry, mining industry, biomedical engineering, environmental engineering, agricultural water-soil engineering, etc.. The internal flow of hydraulic machinery is extremely complex, and its characteristics can be summarized as high Reynolds number, multi-scales, inhomogeneous and vortex-dominant unsteady turbulence which interact with the rotating dynamic boundary（rotor blade）. Based on the analysis of the internal flow characteristics of hydraulic machinery, the author and his research team successively proposed a rotation correction model, a curvature corrected filter-based model, a scalable detached eddy simulation method, and a non-linear hybrid RANS/LES turbulence model to capture unsteady flow structures and then predict hydraulic performance and dynamic characteristics more accurately. According to the analysis on the internal flow, the corresponding flow control measures were put forward. It was verified by experiments that these methods could significantly improve the hydraulic performance, anti-cavitation performance and dynamic characteristics（pressure pulsation and vibration） of hydraulic machinery in a certain range of operating conditions. In addition, the mechanism how flow control measures influence internal flow was analyzed in depth, aiming at finding a feasible and effective way to improve hydraulic performance, anti-cavitation performance and dynamic characteristics of hydraulic machinery.