Hierarchical level fault detection and diagnosis of ship engine systems
作者：
Highlights：
• A proposed algorithm, HL-FDD, is presented for fault detection of ship engines.
• HL-FDD extracts features using optimal clustering and dimension reduction.
• It can detect and impute sensor errors using the Hampel identifier.
• It generates dynamic threshold models labeling the engine condition through loads.
• It monitors hierarchical levels-the entire system, subsystems, and components.
摘要
•A proposed algorithm, HL-FDD, is presented for fault detection of ship engines.•HL-FDD extracts features using optimal clustering and dimension reduction.•It can detect and impute sensor errors using the Hampel identifier.•It generates dynamic threshold models labeling the engine condition through loads.•It monitors hierarchical levels-the entire system, subsystems, and components.
论文关键词：Optimal hierarchical clustering and dimension reduction,Dynamic thresholds,Domain knowledge,Sensor error,Abnormality labeling,Duel fuel marine engine
论文评审过程：Received 4 February 2022, Revised 26 August 2022, Accepted 8 September 2022, Available online 17 September 2022, Version of Record 1 October 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.118814
