Applying Taguchi Method to Improve Experimental Efficiency of Polyacrylonitrile Fiber Spinning Process

碩士 === 國立虎尾科技大學 === 工業工程與管理研究所 === 102 === Now technology is increasingly developed, experiments in various domains are increasingly complex, thus relatively experimental cost and time used are increasingly getting higher, so the study aims at using Taguchi method applied to experiment for polyacryl...

Full description

Bibliographic Details
Main Authors: Yi-Hung Cheng, 程宜鴻
Other Authors: 楊太宏
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/976z2h
Description
Summary:碩士 === 國立虎尾科技大學 === 工業工程與管理研究所 === 102 === Now technology is increasingly developed, experiments in various domains are increasingly complex, thus relatively experimental cost and time used are increasingly getting higher, so the study aims at using Taguchi method applied to experiment for polyacrylonitrile fiber process. The study takes polyacrylonitrile fiber’s acrylic spinning process as example, by means of setting control factors and levels, to evaluate their importance of effect in acrylic strand process, so that accuracy of experimental prediction can be enhanced and cost can be reduced. With related data of acrylic strand process provided by polyacrylonitrile fiber manufacturing plant, and further conducts small range experiment with evaluated experimental combinations, applies Taguchi orthogonal table to evaluate their level of effect in factors affecting quality features, and further conducts small range experiment with evaluated experimental combinations, to find out possible preferred parameter combinations in the process, thereby with less experiment times the objective of reducing experimental cost can be achieved. In acrylic spinning process, the study considers effects of solvent concentration, dope temperature, and flow rate toward raw spinning intensity. The study adopts L18(21×32) orthogonal table to conduct experiment, with factor reaction analysis to observe preferred combinations, and conducts predictions with such combinations, and with experimental combinations from the predictions and interaction analysis, small range experiment is conducted, thus possible preferred parameter combinations in process are found. From experimental results, it is found that the experimental results of 32 times experiments are equivalent to that of 92 times, so at least 1080 hours of experimental time is saved, and more than 60 times of experiments costs are saved. From the experiment, it is found that Taguchi method is capable of effectively enhancing experimental efficiency, reducing experiment cost, and thereby it can be provided for reference in experiments of other processes in synthetic fiber processing plants.