Abstract

Based on the rectifying conduction principle of the Tesla valve, a self-pumping hydrodynamic mechanical seal with Tesla valve-shaped face grooves was proposed, and its corresponding computational model was established. Numerical simulations were conducted to investigate the effects of the Tesla valve diversion angle and valve clearance on the sealing performance of the proposed structure. Taking the leakage rate and liquid film stiffness as the target performance indices, a predictive model was developed by combining uniform experimental design with multiple regression analysis. Subsequently, the NSGA-II (Non-dominated Sorting Genetic Algorithm II) genetic algorithm was employed for bi-objective optimization to obtain the Pareto-optimal solution set, and the TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method was further applied to identify the optimal combination of structural parameters under specified weighting coefficients. The results indicate that the leakage rate is not significantly affected by variations in the diversion angle or valve clearance, whereas the liquid film stiffness increases with increasing diversion angle and decreases with increasing valve clearance. Multi-objective optimization successfully identified an optimal parameter combination that improves the overall sealing performance of the proposed structure. This study provides a novel perspective and theoretical basis for innovation in face structure and for the performance optimization of self-pumping mechanical seals.

IPC Classification

Keywords