Fault Detection Filter Design for a Class of Linear Time-invariable Systems
Shi Song-jiao
Abstract
Shi Song-jiao
Abstract
A new full order detection filter for a class of time-invariable systems is proposed. It obtains a detection filter for the original system by augmenting the original system and designing a new one. This filter can asymptotically detect m (can be increased to n) faults that occur simultaneously with the initial state estimations error. The relation between the detection threshold, designing parameters and the performance of the filter is analysed. The rules for the choice of the parameters and threshold in the design procedure are also given. The simulation result indicates that this detection filter has good detection capability.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A new full order detection filter for a class of time-invariable systems is proposed. It obtains a detection filter for the original system by augmenting the original system and designing a new one. This filter can asymptotically detect m (can be increased to n) faults that occur simultaneously with the initial state estimations error. The relation between the detection threshold, designing parameters and the performance of the filter is analysed. The rules for the choice of the parameters and threshold in the design procedure are also given. The simulation result indicates that this detection filter has good detection capability.
Key concepts: Filter (signal processing), Filter design, Fault detection and isolation, Control theory (sociology), Computer science, Class (philosophy), Detection threshold, Relation (database)