Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Metal additive manufacturing forming size real-time prediction method based on deep learning

A deep learning and metal additive technology, applied in the field of additive manufacturing, can solve problems such as large error in prediction value, interference of image data, and difficulty in accurate processing, and achieve the effect of improving efficiency and accuracy

Active Publication Date: 2019-11-22
SICHUAN UNIV
View PDF7 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the problems faced in the prior art, for example: there is interference in the collected image data, and it is difficult to process accurately; it can be processed in a specific situation, but the generalization performance is poor, resulting in complicated situations. The predicted value has a large error and low accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Metal additive manufacturing forming size real-time prediction method based on deep learning
  • Metal additive manufacturing forming size real-time prediction method based on deep learning
  • Metal additive manufacturing forming size real-time prediction method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] Such as figure 1 As shown, the real-time prediction method of metal additive manufacturing forming size based on deep learning includes the following steps:

[0030] S1: Continuously collect molten pool images in a certain time sequence, use part of the continuous molten pool images to establish a training data set, and similarly, use part of the continuous molten pool images to establish a test data set;

[0031] S2: Establish a deep learning convolutional neural network model, set the corresponding model parameters, including the number of network layers and activation functions; the framework of the network model is Resnet101;

[0032] Such as Figure 5 As shown, the model can be divided into input module, feature extraction module, and decision-making layer module; the input module is the molten pool image, and the feature extraction module is mainly composed of convolutional layer, batch normalization layer, and activation layer. The layer module includes average...

Embodiment 2

[0048] Such as image 3 As shown, the real-time prediction system for the forming size of laser metal additive manufacturing based on deep learning includes a printing table, an image acquisition device, a human-computer interaction device, a display, and a host computer. The host is electrically connected, and the image acquisition device is installed above the printing workbench; the image acquisition device is used to continuously collect images of the molten pool in a certain time sequence and transmit the collected images of the molten pool to the Host; the host is used to set up a training data set using part of the continuous melt pool image, and similarly, also use part of the continuous melt pool image to build a test data set; set up a deep learning convolutional neural network model, and set up a corresponding model Parameters, including the number of network layers and activation functions; the framework of the network model is Resnet101, input the molten pool imag...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a laser metal additive manufacturing forming precision prediction system based on deep learning, which comprises a printing workbench, an image acquisition device, a man-machine interaction device, a display and a host. The image acquisition device, the man-machine interaction device and the display are electrically connected with the host. Molten pool images and temperature images are continuously collected in a certain time sequence; firstly, normalization processing is carried out on an effective molten pool image and an effective temperature image; the parameters ofthe picture size and the pixel size of the molten pool image are kept consistent, other irrelevant features are eliminated during training of the deep learning convolutional neural network model, only key features are trained, and the method has the advantage that the training efficiency of the deep learning convolutional neural network model is improved; and the deep learning convolutional neural network model is adopted to predict the single-channel forming width, so that the precision of parameters can be effectively improved.

Description

technical field [0001] The invention belongs to the technical field of additive manufacturing, and in particular relates to a method for real-time prediction of forming dimensions of metal additive manufacturing based on deep learning. Background technique [0002] The size of single-pass forming, such as forming width, forming height and other parameters are important factors affecting the quality of additive manufacturing, and the characteristics of the molten pool are the most direct factors affecting the forming quality. Therefore, it is of great significance to study the changes of molten pool characteristics in the process of additive manufacturing and realize the control of certain parameters of the molten pool to ensure the quality of additive manufacturing. An important part of intelligence. In recent years, with the development of computer vision technology, it has become an important research direction of additive manufacturing technology to directly observe the ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/00G06T7/70G06N3/04G06N3/08G01B11/00
CPCG06T7/0004G06T7/70G06N3/084G01B11/00G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/30108G06N3/045
Inventor 殷鸣向枭谢罗峰殷国富颜虎刘浩浩李家勇
Owner SICHUAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products