Using Neural Networks and Immune Algorithms to Find the Optimal Parameters for IC Wire Bonding Processes

• Main Authors: Hung-Zhi Chang, 張鴻志
• Other Authors: Tung-Hsu (Tony) Hou
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/87731192832536559968

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 93 === The wire bonding process is the key process in an IC chip-package. It is an urgent problem for IC chip-package industry to improve the wire bonding process capability. In this study, an application of artificial neural networks (ANN) and artificial immune systems (AIS) is proposed to optimize parameters in the wire bonding process in order to achieve highly level performance and quality. In this research, the algorithm of AIS with memory cell and suppressor cell mechanisms is developed, and two new algorithms are proposed：Multiple-Objectives Immume Algorithms(MOIA) and Artifitial Immume System Algotithms(AIS). A back propagation ANN is used to establish the nonlinear multivariate relationships between wire boning parameters and responses. Based on the non-dominated solution found by MOIA with the best parameter setting, the two indices, Error Ratio and Spread, can be used as metrics to measure the performance of MOIA searching the Pareto-optimal-front. Then a Taguchi orthogonal method is applied to identify the critical parameters of AIS. Finally, the MOIA and AIS are applied to find the most desired parameter settings by using the output of ANN as the affinity measure. A comparison between the proposed AIS and a genetic algorithm is conducted in this study. The comparison shows that the searching quality of the proposed AIS is more effective than the GA in finding the optimal wire bonding process parameters. The results shows the MOIA can precisely find the Pareto-optimal-front satisfying multiple objectives, and AIS can find the best manufacturing parameters which can satisfy the single objective limit