Answers are grounded in this paper and cite their sources.
This article presents our view of why machine learning (ML) has not been able to find its way to commercial products in the field of protection. All the methods we have reviewed in the literature essentially propose some ML-based technique to replace a legacy protection scheme. The results are substantiated by training and testing ML models through simulated data. Legacy methods, on the other hand, are developed using mathematical models that capture the physics of disturbances and faults.
No takes yet. Share an insight, caveat, or question.
Wischkaemper et al. (2021) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: