Intelligent inspection and maintenance method and system for AI-enabled industrial equipment

Abstract

The application discloses an AI-enabled industrial equipment intelligent inspection and maintenance method and system, and relates to the technical field of industrial operation and maintenance. The method comprises the following steps: acquiring a current transient energy sequence of a current operation cycle of a device motor; selecting an optimal window length adaptive to the current operation cycle to slide and intercept the current transient energy sequence to obtain a plurality of analysis windows; quantifying the intensity of transient energy change in each analysis window to obtain an energy fluctuation feature sequence of the current operation cycle; selecting a cut-off point from the energy fluctuation feature sequence to obtain a waveform change feature sequence of the current operation cycle; acquiring a waveform change feature sequence of a historical normal sample, calculating a similarity distance corresponding to the current operation cycle; constructing an abnormality judgment mechanism, and judging the device state based on the similarity distance corresponding to the current operation cycle. The application can more accurately detect early hidden faults of the device, and reduce the false positive rate and the false negative rate of fault diagnosis.

Images