Search

Facing the challenges of production management of manufacturing enterprises and the demand for intelligent control of manufacturing system abnormalities, this paper proposes a production abnormal event diagnosis method based on digital twin technology to achieve intelligent tracing of the causes of production abnormal events and improve the problems of poor timeliness and lack of feedback mechanism for the diagnosis of production abnormal events of complex products. First, a complex product production abnormal event diagnosis model is constructed, in which physical production workshop, virtual production workshop, workshop twin data platform and abnormal event diagnosis service system work together. Second, the joint method of Bayesian network (BN) and C5.0 decision tree algorithm (C5.0) is used to achieve the diagnosis of production abnormal events. Finally, the feasibility and effectiveness of the scheme are verified by the arithmetic example of key stations in the bogie production workshop. The diagnostic accuracy of 92.49% of the BN–C5.0 model is higher than that of the conventional C5.0 decision tree model with 79.94% accuracy. The proposed method and mechanism provide a reference model for the application of digital twin to the diagnosis of abnormal production events. [ABSTRACT FROM AUTHOR]

Fat‐tree topologies are widely used in interconnect network designs for parallel supercomputers. In the classic fat‐tree, compute nodes are connected to leaf stage switches by links. Given a large number of compute nodes, many switches and links are required, resulting in high hardware costs. To solve this problem, this paper proposes two hybrid topologies, k$$ k $$‐Cube k$$ k $$‐Ary n$$ n $$‐Tree (CAT) and Mirrored k$$ k $$‐Cube k$$ k $$‐Ary n$$ n $$‐Tree (MiCAT), based on fat‐tree and hypercube. Instead of connecting k$$ k $$ compute nodes directly to a leaf switch, we connect a k$$ k $$‐cube to the switch, and each switch in the k$$ k $$‐cube part connects k$$ k $$ compute nodes. That is, this k$$ k $$‐cube consists of 2k−1$$ {2}^k-1 $$ switches and k(2k−1)$$ k\left({2}^k-1\right) $$ compute nodes. We give the shortest path routing algorithms and evaluate the path diversity, cost, performance, and average packet latency of CAT and MiCAT. The results show that CAT and MiCAT can save up to 87%$$ 87\% $$ switches and 80%$$ 80\% $$ links in a large‐scale parallel system, k=n=8$$ k=n=8 $$ for example, compared to fat‐trees, and meanwhile, both CAT and MiCAT have higher path diversities than fat‐trees. [ABSTRACT FROM AUTHOR]