1. Faulty processor identification for a multiprocessor system under the Malek model using an improved binary bat algorithm.
- Author
-
Gui, Weixia, Pan, Fulai, Zhu, Dandan, and Li, Feng
- Subjects
- *
SYSTEM identification , *ALGORITHMS , *METAHEURISTIC algorithms , *TRANSFER functions , *MULTIPROCESSORS , *FAULT diagnosis - Abstract
A multiprocessor system should be able to identify and eliminate faults in time to avoid the paralysis of a whole system. This paper proposes an improved binary bat algorithm to identify faulty processors in a multiprocessor system. Compared with most existing works based on metaheuristic algorithms, the proposed algorithm employs a random initial population and does not require transfer functions. The exclusive-OR operation in the velocity equation is used to measure the distance between two individuals in binary space. To improve population diversity and avoid local optima, the mutation operator is integrated into the position update equation. A new local search strategy is proposed to strengthen the ability of local search in binary space. Experimental results show that the proposed algorithm based on the Malek model can maintain approximately 100 % diagnostic accuracy in a small random initial population with fewer iterations and less CPU running time. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF