Extract Technology for Defect Voxels in Casting Simulation Data

Zhang Zhaochuang; Sun fei; Liao Dunming; Chen Tao; Tang Yulong; Cao Liu

doi:10.15980/j.tzzz.2017.02.008

Views : 0 下载量: 64 CSCD: 0

Export
Share
Collection
Album

Extract Technology for Defect Voxels in Casting Simulation Data
Vol. 37, Issue 2, Pages: 144-147(2017)
Published： 2017 ，
DOI： 10.15980/j.tzzz.2017.02.008

扫描看全文

[1]张兆创,孙飞,廖敦明,陈涛,唐玉龙,曹流.铸造模拟数据中的缺陷体素单元标识技术[J].特种铸造及有色合金,2017,37(02):144-147.

Zhang Zhaochuang, Sun fei, Liao Dunming, et al. Extract Technology for Defect Voxels in Casting Simulation Data. [J]. Special Casting & Nonferrous Alloys 37(2):144-147(2017)
[1]张兆创,孙飞,廖敦明,陈涛,唐玉龙,曹流.铸造模拟数据中的缺陷体素单元标识技术[J].特种铸造及有色合金,2017,37(02):144-147. DOI： 10.15980/j.tzzz.2017.02.008.

Zhang Zhaochuang, Sun fei, Liao Dunming, et al. Extract Technology for Defect Voxels in Casting Simulation Data. [J]. Special Casting & Nonferrous Alloys 37(2):144-147(2017) DOI： 10.15980/j.tzzz.2017.02.008.

摘要

为了能够让铸造模拟数据更好地适应机器学习的需要

将经验公式和人工分析两种方法一同引入到铸造缺陷预测分析中

在直接体渲染技术基础上

提出了针对铸造模拟数据中的缺陷体素单元的标识技术。实际案例分析表明

这种标识技术可以有效并准确地辅助分析人员将缺陷体素进行标识。凭借标识技术

分析人员观察分析模拟数据并对特定体素进行标识

以此将长期实践经验保存下来。

Abstract

In order to make the simulation data more available for machine learning

two main analysis methods

such as empirical formula and manual analysis

were combined in defect prediction.The extract technology of the casting voxel which is based on the direct volume rendering technology is presenetd.The practical case shows that the method can extract effectively and accurately and save the defect content.In this way

the analysis experience was saved

and therefore

all these saved experience will become the basic data for the following research.