Part attribute matrix automatic generation method based on deep learning

Abstract

The invention discloses a part attribute matrix automatic generation method based on deep learning, and the method comprises the following steps: A, obtaining part information of a product serving asa sample, and creating a part dictionary; B, creating numerical mapping of parts; C, defining the size of an attribute matrix E; D, establishing a part sequence model according to the structure of thesample product; E, setting a fixed sliding window, and dividing a part sequence model to form a training sample set D; F, constructing a neural network structure, and determining an input layer, a hidden layer and an output layer of the network; G, training the related samples; and H, using the attribute matrix E. The invention provides a part attribute matrix automatic generation method based ondeep learning. By means of the method, the attribute matrix can be automatically obtained without manually marking a large number of attributes of a large number of components one by one.

Description

technical field [0001] The invention relates to the technical field of intelligent manufacturing, in particular to a method for automatically generating component attribute matrices based on deep learning. Background technique [0002] At present, in the field of intelligent manufacturing, when using data mining or artificial intelligence to analyze parts, it is often easy to lose their relevance. And enhancing the correlation of parts in data mining can be very effective in understanding the similarity between parts. In product design, accurate and effective component similarity calculation can effectively recommend component selection, so it can improve the efficiency of product intelligent design; in processing process design, because it can greatly improve process resources and process parameters, etc. The reusability of process information can greatly improve the efficiency of intelligent processing process design; in the process of assembly process design, because it ...

AI Technical Summary

Problems solved by technology

[0005] However, for the labeling of attribute information, it often requires a lot of manual operation

Moreover, manually marking and assigning the attribute information of parts is a very resource-intensive work

Images