Chair: Dr. Guicai ZHANG (United Technologies Research Center (China) Ltd)
Deep Learning based Virtual Metrology and Yield Prediction in Semiconductor Manufacturing Processes Prof. Myong Kee (MK) JEONG, Jeongsub CHOI, Youngdoo SON and Jihoon KANG (Rutgers University, Dongguk University, Samsung SDS)
We present a deep learning based supervised autoencoder to extract meaningful features from massive in-line sensor functional signals of semiconductor manufacturing processes. Based on those extract features, we build the virtual metrology model to predict important quality characteristics of the process and the yield prediction model. A real-life case is studied in this work, and the empirical results show that the proposed model outperforms general approaches for the predictions using signal data.
Experimental Study of Dynamic Strain for Gear Tooth using Fiber Bragg Gratings and Piezoelectric Strain Sensors Dr. Liu HONG, Yongzhi QU, Xixin JIANG, Miao HE, David HE, Yuegang TAN and Zude ZHOU (Wuhan University of Technology, University of Illinois at Chicago)
It has always been a critical task to understand gear dynamics for gear design and condition monitoring. Many gear models have been proposed to simulate gear meshing dynamics. However, most of the theoretical models are based on simplified gear structure and may contain approximation errors. Direct measuring of gear strain is important to gear design validation, load analysis, reliability assessment, gear condition monitoring, etc. Most of the existing studies of tooth strain measurements are performed under static load condition. In this paper, we investigate new measuring techniques of using fiber Bragg grating (FBG) sensor and piezoelectric strain for gear dynamic strain measurement. We conduct gear dynamic strain measurement under both low speed and normal speed condition on an industrial gearbox with relatively small module gears. Multi-combinations of speed and load conditions of the gearbox are tested and the results are discussed and analyzed. We analyze multiple factors that affect the tooth root stress, including speed, load, extended tooth meshing, etc. It is found that under low operation speed range, the tooth root strain is mainly determined by the torque, while in the mediate to high speed range, the tooth root strain is jointly affected by speed and torque. Extended tooth contact is shown in the measurement results with strong evidence. It conforms with earlier founding that the transmission error and dynamic load factor are overestimated while the operation smoothness are underestimated for spur gear under heavy load. The measured stains are also compared with numerical simulation.
System Component Degradation: Filter Clogging in a UAV Fuel System Prof. Ian JENNIONS, Zakwan SKAF and Omer EKER (Cranfield University, Artesis Technology Systems A.S.)
The filtration of possible contaminant is an essential part of many engineering processes in industry. Clogging of the filtration medium is one of the primary failure modes in many application areas leading to reduced performance and efficiency. Imitation of real life clogging scenarios in laboratory conditions is not an easy task to perform, but is demonstrated here, with the profiles obtained being injected into a fuel system rig. This paper shows generic results from two benchmark rigs. One is a fuel system laboratory test-bed representing an Unmanned Aerial Vehicle (UAV) fuel system and its associated electrical power supply, control system and sensing capabilities. It is specifically designed in order to replicate a number of component degradation faults with a high degree of accuracy and repeatability. The second is a purpose built filter clogging rig designed to give quality results to aid the development of prognostic algorithms. This paper’s contribution is to show results from the filter clogging rig and derive a transfer function, the relationship between filter clogging pressures and the fuel system valve openings, to enable the fuel system rig to operate as if the clogging filter were part of the system. The results show that the local pressure drop obtained from the fuel rig can be made to closely match the pressure drop levels from the filter clogging rig. This opens up examination of the effects of filter clogging on the full fuel rig system, providing data for future system prognostic work.