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

3D printing error compensation method, system and device based on neural network

A 3D printing and neural network technology, applied in the field of 3D printing, can solve the problems of difficulty in the construction of new model error compensation, and achieve the effect of improving 3D printing accuracy and accurate error compensation

Active Publication Date: 2020-06-16
INST OF AUTOMATION CHINESE ACAD OF SCI +1
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem of difficulty in constructing new model error compensation during the 3D printing process, the first aspect of the present invention proposes a 3D printing error compensation method based on a neural network. Good deformation network / inverse deformation network, which compensates the input model and performs 3D printing based on the compensated model;

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
  • 3D printing error compensation method, system and device based on neural network
  • 3D printing error compensation method, system and device based on neural network
  • 3D printing error compensation method, system and device based on neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0065] Embodiment 1: An embodiment of a neural network-based 3D printing error compensation method

[0066] In this embodiment, the inverse deformation network is a neural network constructed based on the 3D printing inverse deformation function. Compensate the input model through the trained neural network-based inverse deformation network, generate a compensated model, and perform 3D printing based on the compensated model. The inverse deformation network constructed by the inverse deformation function can directly use the model to be printed as input after training, and the output is the corrected model, which can be printed directly.

[0067] The loss function L used for inverse deformation network training in this embodiment is shown in formula (1). The output model of the printed model sample in the training sample is used as the expected output model through the inverse deformation network; in this embodiment, the printed model sample is used as the real input model du...

Embodiment 2

[0113] Embodiment 2: Embodiment of the second neural network-based 3D printing error compensation method

[0114] The difference between this embodiment and Embodiment 1 is:

[0115] The deformation network is constructed according to the 3D printing deformation function; the output model of the pre-printed model samples in the training samples through the deformation network is used as the expected output model. During the training of the deformable network in this embodiment, the model sample before printing is used as the real input model, and the model sample after printing is used as the real output model of the network.

[0116] The input model sample of each training sample is the digital model to be printed, and the output model sample is the digital model obtained from the corresponding 3D printed physical model.

[0117] After the deformation network constructed by the deformation function is trained, the error compensation data of the model data to be printed is ob...

Embodiment 3

[0119] Embodiment 3: Embodiment of the third neural network-based 3D printing error compensation method

[0120] The difference between this embodiment and Embodiment 1 or Embodiment 2 is that: the training samples of the deformable network / inverse deformable network are two-dimensional slice data of a three-dimensional model.

[0121] The reason for operating from 2D data is that 3D data is more difficult to obtain, and 2D data is relatively easier to obtain and equally effective for compensating errors. The specific operation is to slice the 3D model. The slicing operation here is to use the software to perform slicing operations in a direction that is horizontal to the model, and to reduce the dimension of the 3D model. Such as image 3 Shown are the three-dimensional model of the tooth crown, a screenshot of a certain layer of the cut three-dimensional model of the tooth crown, and a two-dimensional image of the layer after sectioning.

[0122] Other parts of the conten...

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 belongs to the field of 3D printing, and specifically relates to a neural network-based 3D printing error compensation method, system, and device, aiming to solve the problem of difficult model error compensation during the 3D printing process. The present invention combines the neural network in the field of artificial intelligence to predict and compensate the errors produced by additive manufacturing, obtain data through three-dimensional scanning and other technologies, and then use the neural network to learn the deformation function in 3D printing and complete the prediction, or learn the reverse deformation function and direct compensation. It is easier to compensate the error of the new model through the present invention, and the accuracy of model printing based on the same hardware is improved, or the requirements for the hardware can be reduced in order to achieve the same accuracy.

Description

technical field [0001] The invention belongs to the field of 3D printing, and in particular relates to a neural network-based 3D printing error compensation method, system and device. Background technique [0002] As a typical cyber-physical system (Cyber-Physical System, CPS), 3D printing has developed very rapidly in recent years. There is a strong demand for mass customization, but the accuracy of 3D printing today still cannot achieve the best results for some printing needs, such as printing dental crowns. The accuracy of 3D printed objects is generally lower compared to traditional methods. Today, the positioning accuracy of commercial 3D printers is at the level of tens or hundreds of microns, and the error of printed objects is generally larger than this. The reason is that in the 3D printing process, there are usually heating and cooling, bonding and polymerization processes, which will cause shrinkage and curling of the printed object. Secondly, with the develop...

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
Patent Type & Authority Patents(China)
IPC IPC(8): B29C64/386B33Y50/00G06N3/04G06N3/08G06T5/00G06T17/20
CPCB29C64/386B33Y50/00G06N3/08G06T5/00G06T2210/41G06T19/20G06T2219/2021G06N3/048G06N3/045G05B19/404G05B19/4099G05B2219/49023
Inventor 沈震熊刚李雨勍高行谢怿赵美华郭超商秀芹董西松武征鹏万力王飞跃
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com