A Digital Twin-Based Approach for Quality Control and Optimization of Complex Product Assembly

Author(s):  
Yuanye Ma ◽  
Hang Zhou ◽  
Honghong He ◽  
Guotao Jiao ◽  
Sha Wei
2021 ◽  
Author(s):  
Xuepeng Guo ◽  
Linyan Liu ◽  
Huifeng Wang ◽  
Tangxiao Yuan

Abstract In order to solve the problem of unmeasurable assembly performance of complex product, the digital twin-driven assembly quality control and prediction of complex product is studied by means of cyber-physical fusion in the assembly workshop. The connotation of digital twin intelligent assembly is introduced, the current research status of complex product assembly quality is compared and analyzed, and three main key technologies for the assembly quality control of complex product are proposed: (1) multidimensional, multi-scale, multidisciplinary modeling and simulating of digital twin-driven assembly; (2) multi-source heterogeneous data collection, sensing and fusion for assembly processes; (3) data-driven decision making, feedback and optimization technology. Finally, the application of digital twin technology in the field of assembly quality control of complex product is prospected.


2022 ◽  
Vol 62 ◽  
pp. 270-285
Author(s):  
Tiago Coito ◽  
Miguel S.E. Martins ◽  
Bernardo Firme ◽  
João Figueiredo ◽  
Susana M. Vieira ◽  
...  

Procedia CIRP ◽  
2021 ◽  
Vol 104 ◽  
pp. 1710-1715
Author(s):  
Stephan Breiter ◽  
Julia C. Arlinghaus

2020 ◽  
Vol 1633 ◽  
pp. 012160
Author(s):  
Xinchun Chen ◽  
Naiqing Yan ◽  
Can Wang ◽  
Peng Ding
Keyword(s):  

2011 ◽  
Vol 403-408 ◽  
pp. 3015-3021
Author(s):  
Ming Zhou Liu ◽  
Zhi Biao Zhao ◽  
Zeng Qiang Jiang ◽  
Mao Gen Ge ◽  
Lin Ling ◽  
...  

For representing the coupling relationships among various correlate elements in the formation process of assembly qualities, the mapping relations between the structure domain and the quality domain was established on the basis of analyzing the hybrid structure assembly. Through classifying the quality characteristics, the quality domain was divided into two sub-sets which are quality control points set and quality attributes set. Defining the relevant elements in the model, such as vertex, arc, etc by the theory of network flow, the coupling relations among various quality control points in the formation process of assembly qualities were reflected. According to the sequence of quality control points entering into assembly environment, their corresponding hierarchy in the model were determined, which presents the timing sexual of the formation process of assembly qualities. At last, a correlation-model between quality attributes and quality control points in the assembly process of complex product based on network flow was established, which would be the theoretical basis for the dynamic tolerance optimization of quality control points. An example of an automobile active gear axle assembly was given to demonstrate the feasibility of the correlation-model.


2021 ◽  
Author(s):  
Xiaowei Guoa

Abstract Product assembly is an important stage in complex product manufacturing. How to intelligently plan the assembly process based on dynamic product and environment information has become an pressing issue needs to be addressed. For this reason, this research has constructed a digital twin assembly system, including virtual and real interactive feedback, data fusion analysis and decision-making iterative optimization modules. In the virtual space, a modified Q-learning algorithm is proposed to solve the path planning problem in product assembly. The proposed algorithm speeds up the convergence speed by adding dynamic reward function, optimizes the initial Q table by introducing knowledge and experience through the case-based reasoning (CBR) algorithm, and prevents entry into the trapped area through the obstacle avoiding method. Finally, take the six-joint robot UR10 as an example to verify the performance of the algorithm in the three-dimensional pathfinding space. The experimental results show that the modified Q-learning algorithm's pathfinding performance is significantly better than the original Q-learning algorithm.


Sign in / Sign up

Export Citation Format

Share Document