MODELING OF THIN FILM PROCESS DATA USING A GENETIC ALGORITHM-OPTIMIZED INITIAL WEIGHT OF BACKPROPAGATION NEURAL NETWORK

Artificial neural network, particularly the backpropagation neural network (BPNN), has been used to construct a prediction model of plasma processes. In building a BPNN model, many training factors are typically involved and the most difficult factor is the initial weight distribution (IWD). In this study, a technique to optimize the IWD effect on BPNN prediction performance is presented. This was accomplished by using genetic algorithm (GA). The experimental data were collected from the etching of silica thin films in a CHF3-CF4 inductively coupled plasma. The etch process was statistically characterized and the etch responses to model include silica etch rate, Al etch rate, Al selectivity, and silica profile angle denoted as anisotropy. The effect of GA parameters (mutation and crossover probabilities) was also evaluated by conducting a 42 full factorial experiment. The performances of GA-BPNN models were compared to those for conventional models. The comparison revealed an improved prediction of GA-BPNN models for all etch responses. The improvement was even more than 15% for all but the Al etch rate data. The proven improvements support the finding that the presented technique is effective in optimizing an IWD effect on BPNN modeling.

[1]  Edward A. Rietman,et al.  Use of neural networks in modeling semiconductor manufacturing processes: an example for plasma etch modeling , 1993 .

[2]  Byungwhan Kim,et al.  Prediction of radio frequency power effect on silicon nitride deposition using a genetic algorithm based neural network , 2006 .

[3]  Goldberg,et al.  Genetic algorithms , 1993, Robust Control Systems with Genetic Algorithms.

[4]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[5]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[6]  Seung-Soo Han,et al.  Using neural network process models to perform PECVD silicon dioxide recipe synthesis via genetic algorithms , 1997 .

[7]  Margaret J. Robertson,et al.  Design and Analysis of Experiments , 2006, Handbook of statistics.

[8]  Gary S. May,et al.  Advantages of plasma etch modeling using neural networks over statistical techniques , 1993 .

[9]  Philipp Slusallek,et al.  Introduction to real-time ray tracing , 2005, SIGGRAPH Courses.

[10]  Peter L. Bartlett,et al.  The Sample Complexity of Pattern Classification with Neural Networks: The Size of the Weights is More Important than the Size of the Network , 1998, IEEE Trans. Inf. Theory.

[11]  Byungwhan Kim,et al.  An optimal neural network plasma model: a case study , 2001 .

[12]  Byungwhan Kim,et al.  Prediction of plasma processes using neural network and genetic algorithm , 2005 .

[13]  Daehyon Kim.  Improving prediction performance of neural networks in pattern classification , 2005, Int. J. Comput. Math..

[14]  Byungwhan Kim,et al.  Temperature effect on deposition rate of silicon nitride films , 2006 .

[15]  Dilip Sarkar,et al.  Randomness in generalization ability: a source to improve it , 1996, IEEE Trans. Neural Networks.

[16]  D. E. Goldberg,et al.  Genetic Algorithms in Search, Optimization & Machine Learning , 1989 .

[17]  G. S. May,et al.  Neurofuzzy modeling of chemical vapor deposition processes , 2000 .

[18]  Zhiping Lin,et al.  Wavelet Packet Multi-layer Perceptron for Chaotic Time Series Prediction: Effects of Weight Initialization , 2001, International Conference on Computational Science.

[19]  M. M. Rai,et al.  Neural Network Modeling of Growth Processes , 2002 .