Dynamic Swapping of Fault- Tolerant Architectures Based on FPGAs for Flexible Manufacturing Systems - Université Polytechnique des Hauts-de-France
Communication Dans Un Congrès Année : 2021

Dynamic Swapping of Fault- Tolerant Architectures Based on FPGAs for Flexible Manufacturing Systems

Gehad Alkady
  • Fonction : Auteur
Ihab Adly
  • Fonction : Auteur
Hany Elsayed
  • Fonction : Auteur
Hani Ragai
  • Fonction : Auteur

Résumé

Nowadays, flexible manufacturing is a vital aspect of factory automation. Focusing on developing countries, it is crucial to increase production line reliability at the lowest extra cost. There are two contributions presented in this paper. First, it is shown how to reduce the cost of a power station and still obtain the same power station reliability as that of a conventional 1-out-of-2 system. Second, the problem of the availability of spare parts (sensors) in developing countries is mitigated by swapping fault-tolerant architectures whenever different quality sensors are used. This can be easily done by housing the production line's controller as well as the sensors' error detection and recovery mechanism in an FPGA and using DPR for swapping mechanism without any costly hardware replacement. As an example, two fault-tolerant architectures are considered: TMR and 4-sift-out. Petri Nets are used to decide which one of the two architectures is more appropriate for a pre-determined level of reliability. Both TMR and Sift-out error detection and recovery mechanisms are synthesized and implemented on a XC7K160TFFV676-1 device.
Fichier non déposé

Dates et versions

hal-03710436 , version 1 (30-06-2022)

Identifiants

Citer

Gehad Alkady, Ihab Adly, Ramez Daoud, Yves Sallez, Hany Elsayed, et al.. Dynamic Swapping of Fault- Tolerant Architectures Based on FPGAs for Flexible Manufacturing Systems. 10th Mediterranean Conference on Embedded Computing, MECO 2021, Jun 2021, Budva, Montenegro. pp.1-5, ⟨10.1109/MECO52532.2021.9460162⟩. ⟨hal-03710436⟩
23 Consultations
0 Téléchargements

Altmetric

Partager

More