Method and system for mining and evaluating extra-high voltage online monitoring data

Abstract

The invention discloses a method for mining and evaluating online monitoring data of an extra-high voltage power device. The method comprises a first step of cleaning online monitored repetitive data; a second step of cleaning online monitored abnormal data based on an association rule; and a third step of carrying out analysis evaluation on the quality of the cleaned online monitored data; judging whether the threshold values of validity, completeness, normativity and redundancy are met; if so, ending; and otherwise, returning to the second step again. The invention also discloses a system for mining and evaluating the online monitoring data of the extra-high voltage power device. The system comprises a repetitive data cleaning unit, an abnormal data cleaning unit and an analysis evaluation unit. The method and the system for mining and evaluating the online monitoring data of the extra-high voltage power device have the beneficial effects that the abnormal data can be distinguished efficiently as the association rule based abnormal data cleaning method is provided, and real-time monitoring and online checking and evaluation of the extra-high voltage online monitoring data are realized as a data quality evaluation system is set up.

Description

technical field [0001] The present invention relates to the technical field of UHV power transmission and transformation, and more specifically, relates to a method and system for mining and evaluating UHV online monitoring data. Background technique [0002] With the continuous advancement of UHV power grid construction and the increasing requirements of users for the safe and stable operation of power systems, power companies continue to expand the breadth and depth of state monitoring of power transmission and transformation equipment, and state monitoring is moving towards high sampling rate, continuous steady-state recording and The trend of large storage has gradually formed big data for UHV power grid status monitoring. How to mine useful information faster and more accurately from these massive online monitoring data has become a hot research field. At present, China's UHV power equipment online monitoring data has a high repetition rate and some prominent problems ...

AI Technical Summary

Problems solved by technology

However, because the transmission cycle set by the CAC device is too short (about 3 minutes), the CAC device will transmit a large amount of duplicate data to the data center, which greatly occupies the storage space of the system

[0005] 2. Abnormal data

[0008] Duplicate and abnormal data have seriously affected the use of data and information mining. Since the development of UHV equipment is in its infancy, there is no complete mining system for its online monitoring data; The abnormal value processing method in the field of online monitoring mostly adopts the method of threshold calibration, which is difficult to distinguish the abnormal value caused by equipment defects, and the abnormal value with small value; the repeated data mostly adopts the method of fixed frequency sampling, which is prone to omission; there is no special purpose for UHV equipment A complete evaluation system for online monitoring data processing quality

Images