A Run-to-Run Profile Control Algorithm for Improving the Flatness of Nano-Scale Products

Abstract

In scaling-up Carbon Nanotubes (CNTs) array manufacturing, the uniformity of CNTs' height, or flatness of array, is critical for the yield of nanodevices fabricated from CNTs, and thus needs to be properly controlled. However, since the flatness of the CNTs array is better characterized by a profile, the conventional run-to-run (R2R) controllers that are designed for a single or multiple quality indicators are not effective in controlling the CNTs array manufacturing process. Therefore, in this work, we first develop a statistical model to characterize the variation of the flatness profile, and then derive a novel R2R profile controller based on a state-space model and Kalman filter to improve the flatness of CNTs array. The performance of the proposed R2R control algorithm is studied and compared with existing controller via simulation studies. Note to Practitioners-In the quality control practice, it should be noted that many aggregated quality indicators or the pass/fail conclusions are, in fact, drawn on the basis of complex measurements. Such measurements convey rich information about the process variability and quality patterns, and is therefore critical to quality improvement.This work focuses on a scaling-up nano-manufacturing process. The quality of the product is characterized by a profile, which contains information about both the location and the flatness of the product. We design a statistical model to characterize product quality, then derive a control algorithm to improve the flatness by adjusting controllable process variables. Such data-driven modeling and control framework is critical to study processes with complex quality metrics and large uncertainties.Index Terms-Nano-manufacturing, run-to-run (R2R) control, statistical quality control.

Elevator Fault Detection Using Profile Extraction and Deep Autoencoder Feature Extraction for Acceleration and Magnetic Signals

“…Profile extraction methods have applied in electric vehicles [14], computer numerical control systems [15] and horizontal planes [16]. Kalman filter [17] is one of the methods being used for profile extraction. However, we have developed an off-line profile extraction algorithm based on low-pass filtering and peak detection to extract elevator start and stop events from sensor data including both acceleration and magnetic signals.…”