Physical constraint-based residual network limited angle combustion field three-dimensional reconstruction method

A 3D reconstruction and limited-angle technology, applied in biological neural network models, neural learning methods, 3D modeling, etc., to achieve high reconstruction accuracy, improve robustness, and solve the effects of gradient explosion

Pending Publication Date: 2022-03-18
NANJING UNIV OF SCI & TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for three-dimensional reconstruction of combustion field with limited angle based on residual network based on physical constraints, to solve the problem of three-dimensional reconstruction of combustion flame emission tomography under limited angle incomplete data projection

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
  • Physical constraint-based residual network limited angle combustion field three-dimensional reconstruction method
  • Physical constraint-based residual network limited angle combustion field three-dimensional reconstruction method
  • Physical constraint-based residual network limited angle combustion field three-dimensional reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0052] This implementation example will illustrate the three-dimensional reconstruction method of the limited-angle combustion field based on the residual network based on the physical constraints proposed by the present invention by performing numerical simulation reconstruction experiments on the flame model.

[0053] The device in this embodiment is composed of 12 CCD cameras arranged in an array, which can effectively save the space of the reconstruction device while performing multi-directional detection. Arrange 12 CCD cameras into three windows, each window has four CCD cameras. The 4 CCD cameras in each window form a window in a 2x2 arrangement, and there are 2 CCD cameras in each column, a total of 2 columns. The two CCD cameras in the same row are distributed along the height direction and fixed on the same fixed frame. The center of the CCD camera located below is parallel to the center of the flame. The distance between the two CCD cameras on the same fixed frame i...

Embodiment 2

[0075] This embodiment will illustrate the three-dimensional reconstruction method of the limited-angle combustion field based on the residual network proposed by the present invention through a real candle flame reconstruction experiment.

[0076] The reconstruction area where the burning candle flame is located is divided into 100 × 100 × 120 grids, and the actual size of each grid is 0.55mm.

[0077] Step 1: The device adopted in this embodiment is the same as the device with three windows and 12 CCD cameras in Embodiment 1. The CCD cameras with three windows and 12 different positions in the device collect 750 frames of continuous flame projection images, each frame It consists of 12 CCD cameras collecting a flame projection image at the same time; extract the position of the characteristic projection of flame emission intensity in 9000 flame projection images, and take the one-dimensional projection data of the same line of projection map corresponding to the position as i...

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 residual network limited angle combustion field three-dimensional reconstruction method based on physical constraints. The method comprises the following steps: taking one-dimensional projection data of a limited angle combustion flame field projection drawing as input data; the physical three-dimensional structure feature distribution of the combustion flame field is used as priori knowledge of a rebuilding result of a constraint flame field, and is introduced into a loss function as a regular term; training the residual network based on the loss function; and an optimized residual network model is established through training, and high-precision three-dimensional reconstruction of a limited-angle combustion flame field is realized. The residual error network model constructed by the invention can realize flame emission tomography three-dimensional reconstruction of a flame combustion field under the condition of limited projection angle, and compared with a reconstruction result of a traditional algebraic iteration algorithm (ART), the reconstruction result obtained by the residual error network model based on physical constraint under the condition of limited angle has higher reconstruction precision; and the method has superiority in calculation time and is higher in reconstruction rate.

Description

technical field [0001] The invention relates to the field of three-dimensional imaging algorithms of combustion flames, in particular to a three-dimensional reconstruction method of a limited-angle combustion field based on a residual network based on physical constraints. Background technique [0002] The three-dimensional measurement and real-time display of the combustion field plays an important role in promoting combustion technology, such as advanced rocket technology and energy utilization technology. To achieve transient 3D parameter reconstruction, the multi-directional projection of the measured dynamic flame must be obtained first. In recent years, the method of flame chemiluminescence tomography (FCT) has aroused widespread interest among researchers, and different forms of multi-directional projection acquisition systems have been established. [0003] Theoretically, only when the number of acquired projections satisfies the Nyquist sampling theorem, can the id...

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 Applications(China)
IPC IPC(8): G06T17/00G06N3/04G06N3/08
CPCG06T17/00G06N3/08G06N3/045
Inventor 郭珍艳荣韶华吴春霞宋旸张玉杰高志山袁群
Owner NANJING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap