A Triple-Magnet Passive Gravity Compensator: Sensitivity-Based Design and Experimental Validation

Abstract Passive magnetic gravity compensation using permanent magnets provides an energy-free solution for counterbalancing gravitational loads; however, the design of such systems typically involves high-dimensional parameter spaces, making conventional optimization methods computationally expensive. This paper proposes a triple-magnet configuration for gravity compensation of linear motion. A simplified design methodology is developed based on sensitivity-guided variable reduction for design optimization, through which the original nine-variable design problem is reduced to a two-stage, one-dimensional search. The proposed method is validated through numerical studies and compared with a full nine-variable genetic algorithm optimization, with results showing that the simplified method achieves an average gravity reduction rate (AGRR) exceeding 93% while reducing computational time by approximately 9-fold. The design is further realized using commercially available magnets and experimentally validated. The proposed approach thus provides a practical concept and efficient framework for the design of passive magnetic gravity compensators for linear motion.