US 11,948,061 B2
Deep auto-encoder for equipment health monitoring and fault detection in semiconductor and display process equipment tools
Heng Hao, Fremont, CA (US); Sreekar Bhaviripudi, Sunnyvale, CA (US); and Shreekant Gayaka, San Jose, CA (US)
Assigned to Applied Materials, Inc., Santa Clara, CA (US)
Filed by Applied Materials, Inc., Santa Clara, CA (US)
Filed on Jan. 6, 2023, as Appl. No. 18/151,156.
Application 18/151,156 is a continuation of application No. 16/545,908, filed on Aug. 20, 2019, granted, now 11,568,198.
Claims priority of provisional application 62/730,477, filed on Sep. 12, 2018.
Prior Publication US 2023/0153574 A1, May 18, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); G05B 19/418 (2006.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06V 10/82 (2022.01)
CPC G06N 3/04 (2013.01) [G05B 19/4189 (2013.01); G06N 3/08 (2013.01); G06V 10/82 (2022.01); G05B 2219/32335 (2013.01); G05B 2219/45031 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for detecting anomalies in a manufacturing process of a substrate comprising:
feeding at least one input time-series trace from one or more sensors associated with one or more manufacturing tools configured to manufacture the substrate to a neural network;
generating, by the neural network, at least one output time-series trace based on the at least one input time-series trace;
calculating an error between a first input time-series trace of the at least one input time-series trace and a corresponding first output time-series trace of the at least one output time-series trace; and
performing a corrective action based on the error.