The invention discloses an equipment intelligent operation and maintenance
prediction system and method based on
deep learning, particularly relates to the field of data communication networks, and is used for solving the problem of inaccurate positioning caused by separation of a prediction conclusion and a detection
receipt. The method comprises the following steps: receiving operation and maintenance detection receipts from a
management plane, extracting a path identifier and a time identifier to generate a measurement path set, establishing a
receipt sequence index, reading a prediction abnormal conclusion, and completing registration calculation of a fragment through significance probability and a
time mark coincidence significance probability in the measurement path set; performing Fisher merge check to obtain a merge saliency result,
cutting to generate a candidate link segment set, expanding port configuration and forwarding relation check consistency in a topology view, outputting a positioning result, retrieving an operation and maintenance
knowledge base based on the positioning result to generate an operation
list, and completing dual consistency check to form an
executable version; and recovering a new operation and maintenance detection
receipt verification positioning result adjustment and maintenance rule item in the
processing window, issuing a rule version index, and writing back the
metadata.