Modified Hopfield Neural Network for Identifying Faults in Symmetric Comparison Models

This paper presents a modified Hopfield neural network for solving the system-level fault diagnosis problem under the symmetric comparison model. The comparison-based self-diagnosis approach assigns tasks to the nodes, and the outcomes from each pair of units performing the same task are compared. The objective is to identify the set faulty of nodes based on the matching and mismatching among the system’s nodes. We consider t-comparison-based diagnosable systems in which at most t nodes can fail permanently at the same time. Results from a thorough simulation study demonstrate the effectiveness of the Hopfield-network-based self-diagnosis algorithm for randomly generated diagnosable systems of different sizes and under various fault scenarios, making it a viable addition or alternative to existing diagnosis algorithms.