Modeling and Optimization of the Blade Structural Parameters for a Turbomolecular Pump

Abstract

The development of high-performance mass spectrometer and vacuum coating technology has placed higher demand on the vacuum level of turbomolecular pumps (TMPs), which are required to possess a greater compression ratio and faster pumping speed. There exists a relation of “as one falls, another rises” between the compression ratio and the pumping speed when traditional improvement methods are used. How to simultaneously increase the compression ratio and pumping speed is a very important question for the high-end turbomolecular pumps. In this study, on the basis of a parallel blade and thin gas aerodynamic model, several types of curved blade are presented to improve the pumping performance of TMPs. The comparison results show that the positive quadratic surface exhibited a better pumping performance than the other curved blades. After that, a hybrid optimization method based on a support vector machine (SVR) and particle swarm optimization (PSO) are proposed to obtain the structural parameters of the rotor blade for the highest pumping speed and maximum compression ratio. The optimization results show that, compared with the parallel blades, the compression single-stage blade row with quadratic surface structure was able to increase the maximum compression ratio by 10.35% and the maximum pumping speed factor by 4.61%. In addition, the intermediate single-stage blade row with quadratic surface structure increased the maximum compression ratio by 9.15% and the maximum pumping speed factor by 2.53%.