Back to results list
Please use this identifier to cite or link to this item:
|Title:||Design and development of an intelligent system for process design of microchip encapsulation|
|Keywords:||Electronic apparatus and appliances -- Plastic embedment|
Hong Kong Polytechnic University -- Dissertations
|Publisher:||The Hong Kong Polytechnic University|
|Abstract:||Microchip encapsulation based on transfer molding is one of the important processes of semiconductor manufacturing. Quality of it is heavily dependent on the transfer mold design, properties of molding compound and parameters setting of transfer molding. In current practice, transfer mold design and parameters setting of transfer molding are done manually in a trial-and-error manner|
In this research, an Intelligent System for process design of Microchip Encapsulation (ISME) has been designed and developed mainly based on case based reasoning (CBR), artificial neural networks (ANNs) and a multiobjective optimization scheme, from which optimal mold design parameters and process parameters setting can be determined. The system consists of two modules, a case based reasoning (CBR) module and a process optimization module. The CBR module is used to determine initial mold design parameters and process parameters setting while the process optimization module is used to determine optimal mold design parameters and process parameters setting. In the design and development of the CBR module, fuzzy set theory was introduced in the case retrieval to improve the quality of retrieved cases and fuzzy regression was employed in the case adaptation to improve the accuracy of adapted solutions. In the design and development of the process optimization module, design of experiments (DOE) techniques, ANNs, multiple regression analysis, and minimax method were employed.
Two validation tests have been conducted. In the first validation test, the initial parameters setting obtained from the fuzzy regression adaptation models was compared with the setting generated by the structured adaptation models. It was found that the setting generated from the fuzzy regression adaptation models is more close to the actual setting. In the second validation test, the quality measures based on the optimal parameters setting obtained from the system was compared with those based on the actual parameters setting with the use of the NN based process model and process simulation. It was found that the predicted quality of the molded packages based on the optimal setting has significant improvement compared with that based on actual setting.
Compared with the process simulation for microchip encapsulation, ISME does not require finite element analysis that could lead to long time for obtaining optimal parameters setting. Implementation of the intelligent system has demonstrated that the time for the determination of optimal mold design parameters and process parameters setting can be greatly reduced and the parameters setting recommended by the system can contribute to the good quality of molded packages.
|Description:||xi, 135,  leaves : ill. ; 30 cm.|
PolyU Library Call No.: [THS] LG51 .H577M ISE 2003 Tong
|Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b17146859_link.htm||For PolyU Users||167 B||HTML||View/Open|
|b17146859_ir.pdf||For All Users (Non-printable)||18.47 MB||Adobe PDF||View/Open|
Checked on Aug 13, 2017
Checked on Aug 13, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.