On-line early warning method for power equipment

Abstract

The invention discloses an online early warning method for power equipment. The online early warning method comprises the steps that a monitoring sequence in a time-time window of the power equipment is obtained; predicted values of monitoring data of all moments in the monitoring sequence are calculated through the prediction algorithm, whether all the predicted values are greater than a preset early warning threshold or not is compared, and early warning information is sent out if any predicted value is greater than the preset early warning threshold; if none of the predicted values are greater than the preset early warning threshold, a monitoring sequence in a history-time window of the power equipment is obtained; predicted values of monitoring data of all moments in the monitoring sequence are calculated through the prediction algorithm, whether all the predicted values are greater than a preset historical early warning threshold or not is compared, and early warning information is sent out if any predicted value is greater than the preset historical early warning threshold. By the implementation of the online early warning method, the limitations caused by the sliding time window and the fixed time window can be eliminated, the slow change of the state of the equipment can be grasped in large time scale, the quick and sudden change of the monitoring sequences can be reflected in time, and therefore the accuracy rate of early warning can be increased.

Description

technical field [0001] The invention relates to the technical field of electric equipment, in particular to an online early warning method for electric equipment. Background technique [0002] In the power system, the correct early warning of the abnormal state of the power equipment is the basis of the condition-based maintenance. For most faults, through real-time monitoring of the characteristic quantity and change trend that can represent the health status of the equipment, it is possible to make a timely and accurate judgment on whether there is an abnormality in the equipment, so as to take measures in advance to eliminate hidden dangers of the equipment and avoid accidents. [0003] The existing monitoring and early warning of power equipment is usually a fixed time window monitoring and early warning model, that is, the detection data obtained by sampling at fixed time intervals is used for analysis and early warning. However, the above early warning model does not f...

AI Technical Summary

Problems solved by technology

[0003] The existing monitoring and early warning of power equipment is usually a fixed time window monitoring and early warning model, that is, the detection data obtained by sampling at fixed time intervals is used for analysis and early warning. However, the above early warning model does not fully consider historical data. For rapidly developing equipment Defects are difficult to find in time, and for long-term and slow-developing equipment defects, it cannot be fully reflected, it is easy to miss detection, and the accuracy of early warning is extremely low

Images