|
Abstract: A mathematical model is built up for the substrate
temperature of the large-aera hot filament chemical vapor deposition
system. A recurrent fuzzy-neural network strategy for the substrate
temperature control system is presented. Then computer simulation and
test study on the system are performed. The results show that the
controller has strong abilities of self-adapting and self-study, can
successfully track the set curve of the substrate temperature and meet
well the need of the temperature control during depositing diamond
films.
Key words: Hot filament chemical vapor deposition Substrate
temperature Fuzzy-neural network Automatic control
CLC No: TP273.5
国家自然科学基金(50075039)和江苏省高校高新技术产业化资助项目.
Received 20040927, received in revised form 20050230
|