This paper presents the optimal design of an exhaust system of a vacuum-compatible air bearing using a genetic algorithm. To use the air bearings in vacuum conditions, the differential exhaust method is adopted to minimize the air leakage, which prevents air from leaking into a vacuum chamber by recovering air through several successive seal stages in advance. Therefore, the design of the differential exhaust system is very important because several design parameters such as the number of seals, diameter and length of an exhaust tube, pumping speed and ultimate pressure of a vacuum pump, seal length and gap(bearing clearance) influence on the air leakage, that is, chamber's degree of vacuum. In this paper, we used a genetic algorithm to optimize the design parameters of the exhaust system of a vacuum-compatible air bearing under the several constraint conditions. The results indicate that chamber's degree of vacuum after optimization improved dramatically compared to the initial design, and that the distribution of the spatial design parameters, such as exhaust tube diameter and seal length, was well achieved, and that technical limit of the pumping speed was well determined.
