Abstract: To solve the logic modeling problem for the formation process of complex mechanical products'assembly quality, assembly area was decomposed into several workstations from the perspective of the formation process of assembly quality, and every workstation was taken as the basic assembly unit to abstract and encapsulate. According to the logical relationships among the workstations, the message transmission network of the whole assembly system was determined, and an Assembly Quality Data Chain based on Object-Oriented Petri Nets (AQDC-OPN) was built to describe the transition and fusion of various kinds of quality characteristics in the assembly process. To promote the ability of autonomy and decision-making, the knowledge rules were adopted. Based on the accumulation law of assembly error, the current quality control points were optimized dynamically which considering of the upstream instantiated quality data, thus the online former feed-forward control of assembly quality and dynamic compensation of error accumulation were achieved. The assembly process of final drive was given to demonstrate the feasibility and effectiveness of the proposed model.

Key words: assembly quality, object-oriented Petri nets, knowledge rules, mechanical products

摘要： 为解决复杂机械产品装配质量形成过程的逻辑建模问题，从装配质量形成过程角度出发，将装配区域划分为若干工作站，将工作站作为基本装配作业单元进行抽象和封装，根据每个工作站对象之间的逻辑关系确定整个系统的消息传递网络，最终建立基于面向对象Petri网的装配质量数据链模型，用以表征装配过程中各类质量特征的变迁和融合。知识规则的引入使模型具有自治性和决策能力，能够结合上游工序质量控制点实例化数据，根据装配误差传递、累积规律对当前工序质量控制点进行动态优化，实现装配质量的前馈控制与误差累积的动态补偿。以主减速器装配过程为例验证了模型的可行性和有效性。

关键词: 装配质量, 面向对像Petri网, 知识规则, 机械产品