MONITORING METHOD AND SUBSYSTEM THAT DETECTS ABNORMAL SYSTEM STATES

摘要

The current application is directed to monitoring subsystems, and monitoring methods incorporated within the monitoring subsystems, that monitor operation of devices and systems in order to identify normal states and to quickly determine when a device or system transitions from a normal state to an abnormal state. The methods and monitoring components to which the current application is directed employ self-organizing maps and moving-average self-organizing maps to both characterize normal system behavior and to identify transitions to abnormal system behaviors.

主权项

1. A system-monitoring method comprising:
providing a state-vector self-organizing map containing units stored in an electronic memory; providing a moving-average self-organizing map aligned with the state-vector self-organizing map and containing elements stored in an electronic memory; and continuously
generating a next set of metric values that defines a current state vector of a system,determining a best-matching unit of the state-vector self-organizing map for the current state vector,using the current state vector to update the best-matching unit and a neighborhood of the best-matching unit within the state-vector self-organizing map and elements of the moving-average self-organizing map corresponding to the best-matching unit and the neighborhood of the best-matching unit within the state-vector self-organizing map, andwhen a scalar value currently associated with the element of the moving-average self-organizing map corresponding to the best-matching unit within the state-vector self-organizing map is below a threshold value,
storing an indication within an electronic memory that the system has likely entered an abnormal state.