Intelligent operation and maintenance alarm filtering method based on multiplatform autonomous prediction and system thereof

Abstract

The invention relates to the maintenance field of operation and maintenance equipment, in particular to an intelligent operation and maintenance alarm filtering method and system based on multi-platform autonomous prediction. The method comprises the following steps: (1) data acquisition integration; (2) data quality inspection; (3) data cleaning; (4) feature engineering; (5) sample collection; (6) model training and parameter optimization; ) model release (8) model use; (9) model feedback and optimization. The system includes a data acquisition and integration module, a data quality inspection module, a data cleaning module, a feature engineering module, a sample sampling module, a model training and parameter optimization module, a model release module, a model alarm filtering module, and a model feedback and optimization module. The invention guarantees the real-time processability of low-level alarm events, and avoids the possibility of potential sudden serious alarm events due to subjective judgment errors of experts and inability to work 24 hours a day.

Description

technical field [0001] The invention relates to the maintenance field of operation and maintenance equipment, in particular to an intelligent operation and maintenance alarm filtering method and system based on multi-platform autonomous prediction. Background technique [0002] The alarm processing of the current mainstream operation and maintenance equipment is still mainly in the direction of subjective judgment, that is, when dealing with lower-level alarm faults, the choice of whether to deal with it is blind (some lower-level alarms are not processed immediately after they occur, It will not affect the normal operation of the system, and the lower-level alarm events may turn into normal later). Therefore, when there is a primary alarm system, it is often necessary to use manual judgment to assist in decision-making whether to deal with low-level alarm events. Specified industry business background: A certain industry field stipulates that the possible alarm levels of a...

AI Technical Summary

Problems solved by technology

Level 5 alarms must be dealt with immediately (already affecting the normal operation of the system), but some level 3 alarms will not affect the progress of normal system business, and there is a possibility that the alarm conditions will no longer be met later

Method 1 requires experts with profound industry experience, which not every company has; experts may also make mistakes in their judgments, and experts cannot guarantee that they can make predictions and judgments 24 hours a day

In addition, the number of low-level alarm events that experts can judge per unit time is also uncertain, and it may not be possible to handle all low-level alarm events

Method 2 requires operation and maintenance personnel to handle all low-level alarm events, which consumes a lot of unnecessary manpower and material resources

In addition, if you wait until all the level 3 alarms are converted to level 5 alarms before processing, although the minimum manpower is used to deal with it, it has already caused irreparable damage to the system

Images