Research On Knowledge Acquisition Technology In The Construction Of Aluminum Electrolysis Knowledge Graph

The core of intelligent manufacturing is based on knowledge system.At present,the knowledge system of aluminum electrolysis industry is not complete,not based on artificial intelligence,and lack of perfect knowledge system.It is of great significance to build knowledge map for promoting intelligent manufacturing.In recent years,the state vigorously promotes the intelligent manufacturing of aluminum electrolysis industry,and the demand for intelligent optimization of aluminum electrolysis industry is increasing.On the one hand,the aluminum electrolysis industry capacity optimization,energy crisis and other issues need to be solved.On the other hand,the intelligent level of management and control of aluminum electrolysis production is not high,the level of knowledge systematization is not strong,and there is a lack of intelligent technical management means.The construction and implementation of aluminum electrolysis knowledge map provides a suitable solution for aluminum electrolysis knowledge storage and industrial intelligent management.Knowledge mapping can clarify the relationship between entities,find the hidden relationship between entities and relationships,promote the intelligence of aluminum electrolysis industry and optimize production.This project takes aluminum electrolysis industry as the background,builds aluminum electrolysis knowledge map,and mainly carries out the following work:(1)This paper analyzes the aluminum electrolysis production technology and knowledge map related technology,and studies the method of extracting entities and the relationship between entities.In the aluminum electrolysis industry entity recognition method,this paper uses bilstm-crf model to extract aluminum electrolysis industry knowledge entity.This paper constructs data set and test set training and test entity recognition model,through continuous training and testing to continuously optimize the model,finally obtains high-quality aluminum electrolysis knowledge entity.For the extraction of the relationship between aluminum electrolysis entities,this paper designs the bilstm attention model.The attention model is added to the bilstm layer for the classification of the relationship between entities.The attention model focuses on the key features in the sentence,obtains the important semantics of the sentence,and completes the task of relationship classification between aluminum electrolysis entities.Finally,the model of entity and relationship between entities is established.(2)The production process of aluminum electrolysis is studied,and the relationship between the extracted entities is stored in the graph database neo4 j.A preliminary knowledge map of aluminum electrolysis is constructed,and the map is analyzed visually.(3)A question answering system of aluminum electrolysis knowledge based on aluminum electrolysis knowledge map is designed and implemented,and the deep learning algorithm is applied to it.