Optimization of propagation in interval constraint networks for tolerance design

Abstract

This paper proposes a hierarchical internal constraint network and interval propagation techniques for automatic tolerance design. The nodes in interval constraint networks represent the entities, the attributes, and the functional requirements of the mechanical design or the constraint functions. The arcs represent the relationships between the entities, the attributes, the functional requirements and the constraint functions. We developed the forward propagation technique for tolerance analysis and the backward propagation technique for tolerance synthesis. In tolerance analysis, given the entity tolerances, the goal is to ensure that the functional requirement tolerances are met. In tolerance synthesis, given the functional requirement tolerances, the goal is to synthesize a new set of entity tolerances. In backward propagation, the minimization of the manufacturing cost is also considered. During backward propagation, the tolerances of entities, which have a smaller impact on manufacturing costs, will be tightened first. Using this mechanism, we ensure the constraints are satisfied and the manufacturing costs are minimized.