Industrial fault knowledge graph establishment method for end-to-end scene

Abstract

The embodiment of the invention discloses an industrial fault knowledge graph establishment method for an end-to-end scene, relates to the field of fault automatic diagnosis based on artificial intelligence, and can establish an industrial fault knowledge graph which is high in quality and large enough to support a follow-up operation reasoning algorithm in a short time. The method comprises the steps that a description text of the industrial equipment fault process is collected and input into a UT-CRF entity naming recognition model, and after an entity naming recognition task is executed, the UT-CRF entity naming recognition model outputs an entity extraction result; extracting a training set from the entity extraction result and training an RCNN entity classification model; the entity extraction result is input into a trained RCNN entity classification model, an entity classification task is executed, and the RCNN entity classification model outputs entity text data with entity category labels; and constructing an industrial fault knowledge graph by using the entity text data with the entity category labels.

Description

technical field [0001] The invention relates to the field of automatic fault diagnosis based on artificial intelligence, in particular to a method for establishing an industrial fault knowledge map for end-to-end scenarios. Background technique [0002] With the development of artificial intelligence technology, the operation and maintenance and fault diagnosis work in the industrial field is gradually developing from traditional manual operation to intelligence and automation. According to the fault work orders accumulated in the past operation and maintenance process, the experience and knowledge extracted can provide data support for intelligent fault diagnosis and operation and maintenance work. As a result, many industrial equipment operation and maintenance and fault diagnosis problems can be transformed into natural language processing problems such as knowledge extraction. At present, the industrial field has accumulated a large amount of fault description data, whi...

AI Technical Summary

Problems solved by technology

[0004] However, in practical applications, it is extremely difficult to construct a high-quality, large-scale knowledge graph. The fully automated construction method is not yet mature, and the quality of the constructed knowledge graph is poor; while the fully manual construction method can guarantee quality, but the efficiency Low, often requires huge manpower and time costs

Images