東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

In-process sensor fusion and data an...

Kim, Jihyun.

Linked to FindBook

Google Book

Amazon

博客來

In-process sensor fusion and data analysis for forging process control and quality improvements.

Record Type:	Electronic resources : Monograph/item
Title/Author:	In-process sensor fusion and data analysis for forging process control and quality improvements./
Author:	Kim, Jihyun.
Description:	102 p.
Notes:	Source: Dissertation Abstracts International, Volume: 65-06, Section: B, page: 3086.
Contained By:	Dissertation Abstracts International65-06B.
Subject:	Engineering, Industrial. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3138198
ISBN:	0496853682

In-process sensor fusion and data analysis for forging process control and quality improvements.
Kim, Jihyun.

In-process sensor fusion and data analysis for forging process control and quality improvements. - 102 p.

Source: Dissertation Abstracts International, Volume: 65-06, Section: B, page: 3086.

Thesis (Ph.D.)--University of Michigan, 2004.

Forging process is very complex with many variables interacting in a complicated manner. Though various in-line sensors are available, only few of them are used for process control and quality improvements. Quality control still relies heavily on part sampling and manual inspection. It is time consuming and also hard to find the root causes of the quality issues. This dissertation is focused on developing methodologies for in-process data fusion for process control and quality improvements. Specifically, three researches have been conducted: (1) Single channel tonnage signal analysis. Principal component analysis is performed on a single channel tonnage signal to obtain principal components which capture the relationship among individual tonnage readings. Nominal process condition identification as well as classification of specific faulty conditions is made possible with data reduction. With this methodology, detailed in-process monitoring can be achieved. (2) Multi-channel tonnage signal analysis. Process condition assessments are made by using principal curves method on multi-channel tonnage signals. This methodology enables concurrent analysis of multi-channel tonnage signals resulting in obtaining more complete assessment of the process condition compared to single channel monitoring. In addition to data reduction, non-stationary signals can be monitored without any prior distribution assumptions. Process classification and detailed understanding of faulty conditions can be acquired as well. (3) Integrated Analysis of Process and Quality Variables . Latent variable modeling is exploited to establish the key relationship between the process and quality variables. Inclusion of overall process information results in thorough assessment of process conditions. Identification of the relative importance of each variable can lead to discovering redundant resources enabling leaner and more focused forging environment.

ISBN: 0496853682Subjects--Topical Terms:

626639
Engineering, Industrial.

In-process sensor fusion and data analysis for forging process control and quality improvements.
LDR:03175nmm 2200289 4500 001 1841688
005 20050915140302.5
008 130614s2004 eng d
020 $a 0496853682
035 $a (UnM)AAI3138198
035 $a AAI3138198
040 $a UnM $c UnM
100 1 $a Kim, Jihyun. $3 1910175
245 1 0 $a In-process sensor fusion and data analysis for forging process control and quality improvements.
300 $a 102 p.
500 $a Source: Dissertation Abstracts International, Volume: 65-06, Section: B, page: 3086.
500 $a Chair: Jianjun Shi.
502 $a Thesis (Ph.D.)--University of Michigan, 2004.
520 $a Forging process is very complex with many variables interacting in a complicated manner. Though various in-line sensors are available, only few of them are used for process control and quality improvements. Quality control still relies heavily on part sampling and manual inspection. It is time consuming and also hard to find the root causes of the quality issues. This dissertation is focused on developing methodologies for in-process data fusion for process control and quality improvements. Specifically, three researches have been conducted: (1) Single channel tonnage signal analysis. Principal component analysis is performed on a single channel tonnage signal to obtain principal components which capture the relationship among individual tonnage readings. Nominal process condition identification as well as classification of specific faulty conditions is made possible with data reduction. With this methodology, detailed in-process monitoring can be achieved. (2) Multi-channel tonnage signal analysis. Process condition assessments are made by using principal curves method on multi-channel tonnage signals. This methodology enables concurrent analysis of multi-channel tonnage signals resulting in obtaining more complete assessment of the process condition compared to single channel monitoring. In addition to data reduction, non-stationary signals can be monitored without any prior distribution assumptions. Process classification and detailed understanding of faulty conditions can be acquired as well. (3) Integrated Analysis of Process and Quality Variables . Latent variable modeling is exploited to establish the key relationship between the process and quality variables. Inclusion of overall process information results in thorough assessment of process conditions. Identification of the relative importance of each variable can lead to discovering redundant resources enabling leaner and more focused forging environment.
520 $a All of the developed methodologies have been validated with real production data through various case studies. A prototype software system has been developed based on the developed methodologies. The research will enable efficient, prompt, and detailed process control and quality improvements for forging process.
590 $a School code: 0127.
650 4 $a Engineering, Industrial. $3 626639
650 4 $a Statistics. $3 517247
690 $a 0546
690 $a 0463
710 2 0 $a University of Michigan. $3 777416
773 0 $t Dissertation Abstracts International $g 65-06B.
790 1 0 $a Shi, Jianjun, $e advisor
790 $a 0127
791 $a Ph.D.
792 $a 2004
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3138198