Mercorelli P. Recent Advances in Intelligent Algorithms for Fault Detection and Diagnosis.
SENSORS (BASEL, SWITZERLAND) 2024;
24:2656. [PMID:
38676272 PMCID:
PMC11053866 DOI:
10.3390/s24082656]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 12/20/2023] [Revised: 04/04/2024] [Accepted: 04/07/2024] [Indexed: 04/28/2024]
Abstract
Fault-finding diagnostics is a model-driven approach that identifies a system's malfunctioning portion. It uses residual generators to identify faults, and various methods like isolation techniques and structural analysis are used. However, diagnostic equipment doesn't measure the remaining signal-to-noise ratio. Residual selection identifies fault-detecting generators. Fault detective diagnostic (FDD) approaches have been investigated and implemented for various industrial processes. However, industrial operations make it difficult to implement FDD techniques. To bridge the gap between theoretical methodologies and implementations, hybrid approaches and intelligent procedures are needed. Future research should focus on improving fault prognosis, allowing for accurate prediction of process failures and avoiding safety hazards. Real-time and comprehensive FDD strategies should be implemented in the age of big data.
Collapse