Knowledge graph relation completion method for numerical control machine tool equipment maintenance

Abstract

The invention discloses a knowledge graph relation completion method for numerical control machine tool equipment maintenance. The knowledge graph relation completion method is characterized in that: constructing a knowledge graph and a candidate relation text, and acquiring a training triple set; enumerating all complete relation paths corresponding to each training triple through a path sorting algorithm; constructing a knowledge graph completion model; selecting any query relationship from the candidate relationship text, inputting the query relationship and all complete relationship paths of the corresponding training triad into a knowledge graph completion model, forming a triad by the query relationship corresponding to the probability score greater than or equal to a set completion threshold and the entity pair, and completing the triad into the knowledge graph. The method has the advantages that the knowledge reasoning capability of the knowledge graph is utilized, a new fault solution is reasoned through an existing fault solution, data are integrated, the problem of data islands is effectively eliminated, the incomplete knowledge graph is complemented, and the blank of graph complementing methods in the field of numerical control machine tool equipment maintenance is filled.

Description

technical field [0001] The invention relates to the field of maintenance of numerical control machine tool equipment, in particular to a knowledge map relation complementing method for the maintenance of numerical control machine tool equipment. Background technique [0002] CNC machine tool equipment is the material and technical basis of production activities. Ensuring the healthy and continuous operation of equipment and reducing the occurrence of risks and failures is the prerequisite for the normal operation of every manufacturing enterprise. However, with the development of economic globalization, the gap between enterprises Competition is constantly increasing, and equipment is becoming more and more mechanized, automated, and intelligent. At the same time, the complexity of equipment system structure is increasing, and the difficulty of equipment fault detection, diagnosis, prediction, and daily comprehensive maintenance is also increasing. If any part of the equipme...

AI Technical Summary

Problems solved by technology

[0002] CNC machine tool equipment is the material and technical basis of production activities. Ensuring the healthy and continuous operation of equipment and reducing the occurrence of risks and failures is the prerequisite for the normal operation of every manufacturing enterprise. However, with the development of economic globalization, the gap between enterprises Competition is constantly increasing, and equipment is becoming more and more mechanized, automated and intelligent. At the same time, the complexity of equipment system structure is increasing, and the difficulty of equipment fault detection, diagnosis, prediction and daily comprehensive maintenance is also increasing. If any part of the equipment system fails, it will affect the normal operation of the entire equipment system

[0003] The existing CNC machine tool equipment maintenance technology adopts the general retrieval method, but due to the scattered and complicated data, the effect obtained by the general retrieval method is very unsatisfactory; and there is actually a correlation between each data, such as equipment and various fault maintenance methods, There is a close relationship between the two, and the general retrieval method does not have a certain correlation with these data, resulting in poor correlation between data and data; in addition, the general retrieval method relies on the experience and knowledge of maintenance personnel and maintenance documents, knowledge is very Scattered, when equipment fails, it is difficult to quickly and effectively carry out maintenance work

Images