Research on the Variation Propagation Model for Complex Mechanical Product Processing

Abstract

The high precision requirements of complex mechanical product (CMP) results in that its processing often contains multiple stages. When part was processed and transmitted in multiple stages, dimensional variation of part will be accumulated, coupling and transformed as well. Since the high precision requirements of the CMP, there are more process parameters impacted the machining errors than general multistage process, and more diverse process parameters should be considered in CMP's processing. To describe machining variation propagation process accurately and effectively, the variation propagation model for CMP processing was proposed. Firstly, the parts and fixtures in CMP processing are described by differential motion vector based on the robot's kinematics principle, and the change of theirs space pose are obtained by homogeneous transformation. Then, the fixture-induced variation, datum-induced variation, machining-induced variation and cutting-tool wear induced variation are analyzed and derived. The machining-induced variation is the dimension deviation produced by cutting force impacting on part's machining surfaces, and the cutting-tool wear induced variation is mainly caused by tool's normal wear. Lastly, the coupling relationship of datum-induced variation, fixture-induced variation, machining-induced variation and cutting-tool wear induced variation of each process stage is described, and the state space model of variation propagation is developed. The relationship of dimensional variation from all work stages and all kinds of variations is explored. By a real case study, feasibility and effectiveness of the model is validated. The model has great potential to be applied in the fault diagnosis, root cause identification of variation, sensor placement optimization, variation simulation, machining parameter and fixture design and optimization in quality control and improvement of complex product.