Digital twin-based process reuse and evaluation approach for smart process planning
Digital twin-based process reuse and evaluation approach for smart process planning
复制标题
基于数字孪生的智能流程规划流程重用和评估方法
DOI：
10.1007/s00170-018-2748-5
复制
发表时间：
2019-02-01
影响因子：
3.4
通讯作者：
Jing, Xuwen
中科院分区：
文献类型：
--
作者：
Liu, Jinfeng;Zhou, Honggen;Jing, Xuwen
随着新一代信息技术的进步，智能工艺规划以更短的时间和更低的成本成为智能工艺规划的重点。基于大数据的多维工艺知识重用与评价是提高企业竞争力的有效策略。然而，如何在加工状态动态变化的情况下对工艺知识进行重用和评价的研究很少。本文提出了一种新的基于数字孪生模型的工艺知识重用和评价方法。首先，引入了包含几何信息和实时工艺设备状态的基于数字对的工艺知识模型，表达了加工工艺规划的目的和要求。其次，基于三层结构及其关联规则构建工艺大数据，用于工艺知识的积累。此外，还提出了场景模型的相似度计算算法，对不匹配的工艺知识进行过滤。为了准确地重用工艺知识，提出了基于实时加工状态和计算置信度的候选知识集工艺可重用性评价方法。最后，将柴油机零部件应用于所开发的原型模块中，验证了所提方法的有效性。该方法可以促进智能工艺设计的发展和应用。
With the advances in new-generation information technologies, smart process planning is becoming the focus for smart process planning with less time and lower cost. Big data-based reusing and evaluating the multi-dimensional process knowledge is widely accepted as an effective strategy for improving competitiveness of enterprises. However, there was little research on how to reuse and evaluate process knowledge with dynamical changing machining status. In this paper, we propose a novel digital twin-based approach for reusing and evaluating process knowledge. First, the digital twin-based process knowledge model which contains the geometric information and real-time process equipment status is introduced to represent the purpose and requirement of machining planning. Second, the process big data is constructed based on the three-layer and its association rules for accumulating process knowledge. Moreover, the similarity calculation algorithm of the scene model is proposed to filter the unmatched process knowledge. For accurately reusing the process knowledge, the process reusability evaluation approach of the candidate knowledge set is presented based on the real-time machining status and the calculated confidence. Finally, the diesel engine parts are applied in the developed prototype module to verify the effectiveness of the proposed method. The proposed method can promote the development and application of the smart process planning.