Flow data prediction and abnormality detection
AssigneeThe Regents of the University of Michigan
InventorsRomesh Saigal, Abdullah O. AlShelahi, Ben Wang, Jeffrey Stephen Choy
Life of this patent
Estimating the term…
Loading patent…
AssigneeThe Regents of the University of Michigan
InventorsRomesh Saigal, Abdullah O. AlShelahi, Ben Wang, Jeffrey Stephen Choy
Estimating the term…
A method for detecting an abnormality in a flow system includes obtaining, with a processor, input flow data for the flow system over a series of time intervals, sequentially processing, with the processor, the input flow data to generate, for each time interval in the series of time intervals, flow feature data, the flow feature data being representative of a plurality of flow parameters for the input flow data at the time interval in the series of time intervals, applying, with the processor, the flow feature data to a machine learning tool, the machine learning tool being configured to provide an assessment, for each time interval in the series of time intervals, of whether the input flow data is indicative of the abnormality being present in a following time interval in the series of time intervals, and providing, with the processor, output data indicative of the assessment.
A method for detecting an abnormality in a flow system, the method comprising:
A system for detecting an abnormality in a flow system, the system comprising:
A method for detecting an abnormality in a flow system, the method comprising:
Dependencies are inferred from the claim text.