Laser metal additive deposition fusion state real-time prediction method and system

Abstract

The invention discloses a laser metal additive deposition fusion state real-time prediction method and system, and the method comprises the following steps: (1), building a real-time prediction initial model, the real-time prediction initial model comprising a backGAN network used for converting a process parameter data set into image data, and a plurality of paths of parallel convolutional neuralnetworks connected with the backGAN network and used for extracting feature data in the image data, a feature fusion layer connected with the multiple paths of parallel convolutional neural networksand used for carrying out feature fusion on the feature data output by the multiple paths of convolutional neural networks, a full connection layer and a classification layer sequentially connected behind the feature fusion layer; (2), training the real-time prediction initial model to form a real-time prediction final model; and (3), inputting a process parameter data set into the real-time prediction final model to predict a laser metal additive deposition fusion state in real time. By constructing the prediction model of the backGAN network and the VGG19 network, the molten pool state can be predicted through process parameters, a neural network with supervised learning is formed, and the cost is low.

Description

technical field [0001] The invention belongs to the technical field of metal additive manufacturing, in particular to a method and system for real-time prediction of fusion state of laser metal additive deposition. Background technique [0002] Due to its easy forming of complex parts and good coating processing, laser metal additive manufacturing technology is widely used in medical, aerospace and defense fields. In the laser metal additive manufacturing technology, the deposition fusion state of a single channel and single layer is an important factor affecting the forming quality of parts, and among the many process parameters that affect the deposition fusion state, laser power, scanning speed, spot diameter, powder feeding The amount and carrier gas velocity are the focus of current research in the field of laser metal additive manufacturing. Therefore, it is of great significance to study the influence of the changes of the above process parameters on the fusion state...

AI Technical Summary

Problems solved by technology

[0003] At present, process parameters such as laser power, scanning speed, spot diameter, powder feeding amount and carrier gas speed are often collected through speed or displacement sensors and background data directly from the industrial computer of the experimental platform, and then a logistic regression model is constructed to realize the deposition analysis. Real-time prediction of the fusion state, or prediction through the state image of the molten pool, this kind of prediction method needs to obtain data from the printer, and predict after processing the data. There must be a time difference in the data transmission process, which will cause deposition fusion. Real-time prediction of state often has poor generalization performance and insufficient robustness

Images