A coal equipment failure early warning system and method based on machine learning

Abstract

The invention discloses a coal equipment failure early warning system and method based on machine learning. The data signal acquisition unit collects real-time data and acquires static data; the database acquires data from the data signal acquisition unit, and organizes the data according to the hierarchical structure of mine equipment , form a device-level data model, and store the device-level data model; the machine learning platform communicates with the data signal acquisition unit and / or database, and can provide the basic machine learning algorithm and operating environment; the machine learning algorithm module passes through the machine learning platform Based on the machine learning basic algorithm and operating environment on the computer, analyze the data and establish the equipment failure early warning analysis model, and judge the equipment failure early warning through the equipment failure early warning analysis model; display the failure early warning judgment results of the machine learning algorithm module through the result display unit . The invention can timely discover the early abnormal situation of the equipment, and issue an early warning message according to the detection result.

Description

technical field [0001] The invention belongs to the technical field of early warning of coal equipment failures, and in particular relates to a coal equipment failure warning system and method based on machine learning. Background technique [0002] Coal equipment is the core tool of coal production. Once a failure occurs, it will not only seriously damage the equipment itself, but also affect the entire production and transportation system, and even endanger the lives of personnel, bringing huge safety hazards and economic losses to coal mine production. Therefore, the research on coal equipment fault warning system is particularly important. [0003] At present, the online fault warning system for coal equipment mainly has the problems of untimely alarm and inaccurate alarm, among which the alarm threshold setting and the single early warning judgment rule are two key problems. In general, the alarm threshold is mostly obtained from the experience of the manufacturer or t...

AI Technical Summary

Problems solved by technology

[0003] At present, the online fault warning system for coal equipment mainly has the problems of untimely and inaccurate alarms, among which the alarm threshold setting and single warning judgment rules are two key problems

In general, the alarm threshold is mostly obtained from the experience of the manufacturer or through the historical data of the equipment. The above two acquisition methods have factors that do not consider the on-site environmental problems and that the online operation equipment is already in a "sick" state, so the collection cannot be guaranteed. The historical data is the state data when the equipment is running normally

In addition, in the actual coal production process, any single fault early warning judgment rule cannot fully represent the fault of the equipment. In the case of fully considering the operating environment of coal equipment, it is also considered that the judgment of equipment fault is easily affected by the production conditions.

Images