Method for building industrial neural network based on graph database

Abstract

The invention discloses a method for building an industrial neural network based on a graph database. The method comprises the following steps: dividing a target industrial manufacturing process intoa plurality of industrial manufacturing stages; carding and obtaining a plurality of operation services contained in each stage; analyzing and obtaining data, constraint conditions and applications associated with each operation service; establishing a corresponding sub-graph database for each operation service; gathering to form a total graph database; configuring a first type of interface associated with an algorithm library in a general graph database; and establishing an industrial neural network comprising a total graph database and an algorithm library. According to the method for building the industrial neural network based on the graph database, various related elements of each stage in the industrial manufacturing field are effectively integrated by combining the graph database technology and the algorithm library, the industrial process and various complex relations contained in the industrial process can be completely and clearly described, and therefore industrial intelligent production can be achieved.

Description

technical field [0001] The invention relates to graph database technology and related technologies of an algorithm library, in particular to a method for building an industrial neural network based on a graph database. Background technique [0002] In the traditional industrial manufacturing field, such as design, production and manufacturing process, start from the demand, transform the demand into each function of the design, and then design from the function. Afterwards, the designed data will be combined with manufacturing for production verification. Various practical problems will arise during the production verification stage. While solving the problems, the design plan will be optimized, and the final plan will be formed after continuous iteration. [0003] In the traditional industrial manufacturing field, the design-to-manufacturing process as mentioned above is relatively long, and it is prone to various problems caused by inaccurate and untimely data transmission...

AI Technical Summary

Problems solved by technology

[0003] In the traditional industrial manufacturing field, the design-to-manufacturing process as mentioned above is relatively long, and it is prone to various problems caused by inaccurate and untimely data transmission, such as complex problems involving design, manufacturing and physical products. Matching issues between different models, docking issues in the whole quality life cycle, etc.

Therefore, the traditional industrial design and manufacturing process is increasingly unable to meet the rapid development of industrial production needs. Therefore, some corresponding solutions have been proposed in related fields, such as gathering massive industrial data from the perspective of data. In the form of the Industrial Internet, a new industrial design, production and manufacturing model is formed to promote the development of the industry

[0004] However, most of the current industrial Internet can only realize the aggregation of local data, but cannot combine multiple stages such as design and manufacturing, manufacturing and testing, design and testing and related data.

As a result, the operation of a process often goes through multiple Internet platforms, while the knowledge in the existing industrial manufacturing field is fragmented and independently distributed, and mostly relies on the experience of technicians, which brings inconvenience to the intelligent decision-making of the industry

Images