Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/2251
Title: An intelligent system for the determination of initial process parameter setting for injection moulding
Authors: Mok, Siu-lung
Keywords: Injection molding of plastics -- Data processing
Artificial intelligence -- Industrial applications
Hong Kong Polytechnic University -- Dissertations
Issue Date: 2000
Publisher: The Hong Kong Polytechnic University
Abstract: The global competition that demands high quality plastic products and short time-to-market has made the current trial and error practice in the determination of initial process parameters for injection moulding become inadequate. According to the nature of the problem in initial process parameter setting for injection moulding, case based reasoning (CBR) is deemed to be a promising technique to handle the experience-based problems. In this research, a hybrid neural network and genetic algorithm (NNGA) approach was introduced to complement the CBR approach in the determination of initial process parameters for injection moulding, from which a Hybrid System for Injection Moulding (HSIM) was developed. In the system, initial process parameters of injection moulding are generated in two attempts. In the first attempt, initial process parameters are generated based on CBR approach. If there is no workable solution to be obtained from the first attempt, the second attempt in the generation of initial process parameters for injection moulding is performed based on hybrid NN-GA approach. HSIM was validated by using a commercial simulation package for injection moulding. Results of the system validation indicate that HSIM can generate a set of initial process parameters for injection moulding that can lead to the production of good quality moulded parts. Implementation of HSIM has also demonstrated that the time for the determination of initial process parameters for injection moulding can be greatly reduced, daily experience of moulding personnel in initial process parameter setting can be captured, and self-learning capability can be facilitated.
Description: xiii, 139, 20 leaves : ill. ; 30 cm.
PolyU Library Call No.: [THS] LG51 .H577M MFG 2000 Mok
URI: http://hdl.handle.net/10397/2251
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
E-thesis_Link.htmFor PolyU Users162 BHTMLView/Open
b15353850.pdfFor All Users (Non-printable)5.34 MBAdobe PDFView/Open
Show full item record

Page view(s)

498
Last Week
2
Last month
Checked on May 21, 2017

Download(s)

455
Checked on May 21, 2017

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.