Multi-scale martensite microstructure aging and damage grading method

A technology of micro-organization and grading method, applied in neural learning methods, image analysis, image data processing and other directions, can solve problems such as difficult to meet effective data analysis

Pending Publication Date: 2020-12-25
CHINA SPECIAL EQUIP INSPECTION & RES INST
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, only research on microstructure aging and damage characteristics of metallographic image data with a fixed ratio in the experimental environment is difficult to meet the needs of effective analysis of such data

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
  • Multi-scale martensite microstructure aging and damage grading method
  • Multi-scale martensite microstructure aging and damage grading method
  • Multi-scale martensite microstructure aging and damage grading method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035]This application is an automatic grading method for martensite damage and aging based on a deep residual network, using the deep learning framework PyTorch convolutional neural network model, and using the image preprocessing library torchvision to preprocess the image.

[0036] Taking 315 500× (magnification 10,000 times) microstructure pictures of iron and steel materials collected in the laboratory as an example, deep learning is used to automatically identify the microstructure of iron and steel materials.

[0037] This application discloses an automatic grading method for martensite damage and aging based on deep learning, which is specifically carried out according to the following steps:

[0038] Step 1. Determine the standard resolution, that is, the magnification a of the steel microstructure to be identified, where 50 figure 2 shown. After being graded by experts for aging damage, the ratings are bound to pictures to build a dataset.

[0039] Step 2. Perform t...

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 multi-scale martensite microstructure aging and damage grading method which comprises the following steps: collecting microstructure pictures of a high-chromium martensite heat-resistant steel material by using a metallographic microscope with a specific magnification factor to construct a data set, and marking a grade label representing aging and damage degrees for eachpicture; reducing all the pictures, and constructing a multi-scale data set containing different resolutions; initializing a neural network by using pre-trained model parameters, and constructing a multi-scale metallographic classification model based on a residual neural network feature pyramid; presetting hyper-parameters of learning rate and iteration times, adopting cross entropy as a loss function, and applying a stochastic gradient descent method to finely adjust the constructed model; through metallographic microscopes with different resolutions, high-chromium martensite heat-resistantsteel microstructure pictures needing to be recognized are obtained, a plurality of small pictures with determined sizes are taken out of the pictures, and the trained model is used for grading. The model trained by the method can be extended to pictures with various resolutions for use.

Description

technical field [0001] The invention relates to the field of aging and damage identification of high-chromium martensitic heat-resistant steel microstructure, in particular to a method for automatic grading of metallographic structure. Background technique [0002] The microstructure characteristics of iron and steel materials are one of the important factors determining the properties of materials. Due to the influence of environment, temperature, pressure and other factors, the microstructure of steel materials often undergoes aging and damage to varying degrees during use, which brings great hidden dangers to safe production. Therefore, how to scientifically and efficiently detect the aging and damage of steel materials has become one of the problems to be solved urgently in theory and practice. In recent years, in the field of thermal power generation, high-chromium martensitic heat-resistant steels represented by P91 steel and P92 steel have been widely used in key pre...

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): G06K9/00G06K9/62G06N3/04G06N3/08G06T3/40G06T7/00
CPCG06N3/084G06T3/4007G06T7/0008G06T2207/10056G06T2207/20081G06T2207/20084G06T2207/30136G06V20/695G06V20/698G06N3/045G06F18/241G06F18/214
Inventor杨旭钱公陈新中徐光明施超段也
OwnerCHINA SPECIAL EQUIP INSPECTION & RES INST