首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0708108-200614


URN etd-0708108-200614 Statistics This thesis had been viewed 1892 times. Download 8 times.
Author Jia-Cheng Shih
Author's Email Address No Public.
Department Mechanical Engineering
Year 2007 Semester 2
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 103
Title DEVELOPMENT OF INTELLIGENT FUZZY
DESIGN IDENTIFICATION SYSTEM ON
PLASTIC INJECTION MOLDED TV BASE
Keyword
  • Best of wisdom
  • genetic algorithms
  • back-propagation networks
  • Finite Element Method
  • fuzzy intention
  • fuzzy intention
  • Finite Element Method
  • back-propagation networks
  • genetic algorithms
  • Best of wisdom
  • Abstract 3 C with the vigorous development of industry, the plastic products have not only widely used in consumer electronics products, and gradually replacing traditional materials as bearing components. Plastic products between appearance and strength, are often highly complex and nonlinear interactions between, this study selected LCD TV's plastic support, considerations to bear the load under the deformation of the quality characteristics of the establishment of vague intention of the design Good system.
    This study by fuzzy AHP of the survey, analysis and the results confirmed the intention of consumers, depending on selected types of support, the identification of LCD-TV related withstand deformation control factor, through the Taguchi method orthogonal experimental planning table To ANSYS simulation of the plastic-bearing capacity after deformation, to understand individual parameters on the quality of the effects of the geometric shape and gain the optimum combination of design and construction through the analysis of data back-propagation network of learning forecasting model ; Further with the quality of forecasting mechanism and genetic algorithms, LCD TV supporting a specific quality and the best quality of the design plan, and to support product development and design process appearance, The development of a " Finite Element Method to establish LCD TV supporting intelligent fuzzy geometry of the system the best intentions."
    Taguchi experiment can effectively reduce the number of simulation and neural network using Taguchi Methods Analysis of the results of the study and forecast effectively within the scope of data, errors are less than 10%. Thus the use of genetic algorithms to search the whole domain got than Taguchi to find the optimal mix optimized the parameters of more than 5%, confirmed the feasibility of this study. Under this study, hope for the future can be used to shape the development of products designed to reduce the development of design time.
    Advisor Committee
  • Ming-Yung Wang - advisor
  • Ching-Chih Tai - co-chair
  • Long-Jyi Yeh - co-chair
  • Tian-Syong Lan - co-chair
  • Files indicate in-campus access only
    Date of Defense 2008-06-24 Date of Submission 2008-07-09


    Browse | Search All Available ETDs