Predictive analysis for remote machine monitoring
View Patent ↗Methods for predicting a machine event are described. In one aspect, a method includes receiving data for current events for the machine, and determining whether the data for the current events is within operational limits. The method also includes, when the data for the current events is within operational limits, determining, using a predetermined model, whether an anomaly exists, and generating an alert including information on the anomaly when the anomaly exists. Systems and machine-readable media are also described.
1. A method for predicting a machine event for a mining machine comprising:
receiving data for current events for the mining machine;
determining whether the data for the current events is within operational limits; when the data for the current events is within operational limits, determining, using a predetermined model, whether an anomaly exists; and
generating an alert comprising information on the anomaly when the anomaly exists,
wherein the data for current events comprises temperature for crowd bearings on a crowd motor drive end and temperature of crowd bearings at a non-driven end for a crowd belt oil the mining machine, and wherein the predetermined model monitors relative temperature between the crowd bearings and provides a cross-correlation between bearing temperatures and tension of the crowd belt to identify the anomaly for the crowd belt tension.
2. The method of claim 1 , wherein the predetermined model identifies the anomaly when bearing temperatures increase more than three standard deviations.
3. The method of claim 1 , including the step of, when the data for the current events is not within operational limits, discarding the data for the current events.
4. The method of claim 1 , wherein determining whether another anomaly exists comprises using a predetermined model to identify a voltage dip of the incoming line voltage that occurs upon actuation of a power bridge of a drive of the drive system of the mining machine.
5. The method of claim 1 , wherein determining whether another an anomaly exists using a predetermined model comprises comparing current events for the mining machine with past events for the mining machine.
6. The method of claim 1 , further comprising transmitting the alert to a user.
7. The method of claim 1 , wherein the alert is transmitted by at least one of a telephone call, voice notification, electronic message, text message, or instant message.