Prediction of chamber leak pattern using time-series neural network

Authors:
Minji Kwon;Byung Chan Park;Byungwhan Kim
Affiliations:
Department of Electronic Engineering, Sejong University, Seoul, Korea;Semisysco Inc., Suwon, Korea;Department of Electronic Engineering, Sejong University, Seoul, Korea
Venue:
ICS'08 Proceedings of the 12th WSEAS international conference on Systems
Year:
2008

Citing 1
Cited 1

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Combining a recurrent neural network and a PID controller for prognostic purpose: a way to improve the accuracy of predictions

WSEAS Transactions on Systems and Control

Quantified Score

Hi-index	0.00

Visualization

Abstract

A leak of plasma chamber should be strictly monitored to maintain process quality as well as device yield. Using an Auto-Correlated Time Series Neural Network (A-NTS), a prediction model of chamber leak was developed. A total of 47 leak patterns were used to construct and test the monitoring efficacy of model. The validation errors for normal and abnormal data ranged between 51.9 and 61.7, and between 126.0 and 163.8, respectively. The validation errors for the abnormal data were more than two times larger than those for the normal data. This clearly indicates that the A-NTS model can accurately detect a leak from the chamber of plasma equipment.