Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process
碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 101 === The rapid development of the human civilization, human dependency and demand for natural resources climbing .However, in the case of natural resource reserves are depleted, humans began to look for other alternative sources of energy to solve these probl...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2013
|
Online Access: | http://ndltd.ncl.edu.tw/handle/19015439897333050066 |
id |
ndltd-TW-101YUNT5031042 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-101YUNT50310422015-10-13T22:57:22Z http://ndltd.ncl.edu.tw/handle/19015439897333050066 Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process 利用自適應基因演算法於太陽能電池網版印刷製程參數最佳化之研究 Yau-shian Huang 黃耀賢 碩士 國立雲林科技大學 工業工程與管理研究所碩士班 101 The rapid development of the human civilization, human dependency and demand for natural resources climbing .However, in the case of natural resource reserves are depleted, humans began to look for other alternative sources of energy to solve these problems. Solar Cells selected as an alternative energy future of the most forward-looking. The Solar Cell Screen Printing Process for one of the process affect the focus of the solar cell conversion efficiency. It’s a complex nonlinear relationship between the control parameters and quality characteristics of the solar cell stencil printing process, It’s a multi-objective optimization problem, and engineers often a trial-and-error method or rule of thumb in the regulation of control machine parameters, correction, but these methods time consuming and there is the risk of error adjustment. This study combined with Back-propagation neural network (BPN) and Genetic Algorithms (GA) search screen printing process in the best combination of parameters. First, the use of back-propagation neural network constructed prediction module, and then the optimal parameters through the following two methods search: 1. Multiple Objective Genetic Algorithm (MOGA) and 2. Adaptive Multiple Objective Genetic Algorithm (AMOGA), Final evaluation Desirability Function of each algorithm. The results show that AMOGA has the best search performance, and better than other methods. Expect AMOGA search the best combination of parameters can provided the case company, and it regulation have a more objective frame of reference. Tung-Hsu Ho 侯東旭 2013 學位論文 ; thesis 68 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 101 === The rapid development of the human civilization, human dependency and demand for natural resources climbing .However, in the case of natural resource reserves are depleted, humans began to look for other alternative sources of energy to solve these problems. Solar Cells selected as an alternative energy future of the most forward-looking. The Solar Cell Screen Printing Process for one of the process affect the focus of the solar cell conversion efficiency. It’s a complex nonlinear relationship between the control parameters and quality characteristics of the solar cell stencil printing process, It’s a multi-objective optimization problem, and engineers often a trial-and-error method or rule of thumb in the regulation of control machine parameters, correction, but these methods time consuming and there is the risk of error adjustment.
This study combined with Back-propagation neural network (BPN) and Genetic Algorithms (GA) search screen printing process in the best combination of parameters. First, the use of back-propagation neural network constructed prediction module, and then the optimal parameters through the following two methods search: 1. Multiple Objective Genetic Algorithm (MOGA) and 2. Adaptive Multiple Objective Genetic Algorithm (AMOGA), Final evaluation Desirability Function of each algorithm.
The results show that AMOGA has the best search performance, and better than other methods. Expect AMOGA search the best combination of parameters can provided the case company, and it regulation have a more objective frame of reference.
|
author2 |
Tung-Hsu Ho |
author_facet |
Tung-Hsu Ho Yau-shian Huang 黃耀賢 |
author |
Yau-shian Huang 黃耀賢 |
spellingShingle |
Yau-shian Huang 黃耀賢 Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
author_sort |
Yau-shian Huang |
title |
Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
title_short |
Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
title_full |
Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
title_fullStr |
Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
title_full_unstemmed |
Using Self-Adaptive Genetic Algorithm to Find the Optimization Manufacturing Parameters for a Solar Cell Screen Printing Process |
title_sort |
using self-adaptive genetic algorithm to find the optimization manufacturing parameters for a solar cell screen printing process |
publishDate |
2013 |
url |
http://ndltd.ncl.edu.tw/handle/19015439897333050066 |
work_keys_str_mv |
AT yaushianhuang usingselfadaptivegeneticalgorithmtofindtheoptimizationmanufacturingparametersforasolarcellscreenprintingprocess AT huángyàoxián usingselfadaptivegeneticalgorithmtofindtheoptimizationmanufacturingparametersforasolarcellscreenprintingprocess AT yaushianhuang lìyòngzìshìyīngjīyīnyǎnsuànfǎyútàiyángnéngdiànchíwǎngbǎnyìnshuāzhìchéngcānshùzuìjiāhuàzhīyánjiū AT huángyàoxián lìyòngzìshìyīngjīyīnyǎnsuànfǎyútàiyángnéngdiànchíwǎngbǎnyìnshuāzhìchéngcānshùzuìjiāhuàzhīyánjiū |
_version_ |
1718082722540290048 |