| 摘要 |
<p>Method 10 of identifying anomalies in a monitored system includes acquiring input data from a plurality of sensors in the monitored system, preprocessing the acquired data to prepare it for modeling leaving a first data subset that feeds 20 into a normal Gaussian mixture model built using normal operating conditions of the monitored system, data is flagged as anomalous by the normal Gaussian mixture model leaving a second data subset that is compared to at least one threshold 22 If the comparison indicates that the second data subset contains anomalies 24, then the second data subset feeds 26,28 into at least one of a set of asset performance Gaussian mixture models which Identify which data contributes to an abnormality in the monitored system leaving a third data subset. Post-processing the third data subset extracts anomalies in the monitored system. The method may use Bayesian networks and influence networks to classify the extracted anomalies.</p> |
