US 12,242,781 B2
Systems and methods for machine learning based fast static thermal solver
Norman Chang, Fremont, CA (US); Hsiming Pan, San Jose, CA (US); Jimin Wen, Pleasanton, CA (US); Deqi Zhu, San Jose, CA (US); Wenbo Xia, Milpitas, CA (US); Akhilesh Kumar, Milpitas, CA (US); Wen-Tze Chuang, Yilan County (TW); En-Cih Yang, New Taipei (TW); Karthik Srinivasan, Cupertino, CA (US); and Ying-Shiun Li, Pleasanton, CA (US)
Assigned to ANSYS, INC., Canonsburg, PA (US)
Filed by ANSYS, INC., Canonsburg, PA (US)
Filed on Nov. 13, 2023, as Appl. No. 18/389,212.
Application 17/664,241 is a division of application No. 16/709,746, filed on Dec. 10, 2019, granted, now 11,366,947, issued on Jun. 21, 2022.
Application 18/389,212 is a continuation of application No. 17/664,241, filed on May 20, 2022, granted, now 11,853,661.
Prior Publication US 2024/0078362 A1, Mar. 7, 2024
Int. Cl. G06F 30/27 (2020.01); G06F 113/18 (2020.01); G06F 119/08 (2020.01)
CPC G06F 30/27 (2020.01) [G06F 2113/18 (2020.01); G06F 2119/08 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A non-transitory machine-readable medium storing executable program instructions which when executed by a data processing system cause the data processing system to perform a method, the method comprising:
selecting a neural network model from a plurality of neural network models, each of the neural network models having been trained with a plurality of inputs and an output derived from thermal simulations to predict a temperature rise of a center tile of a plurality of tiles of a corresponding template based on predefined power levels for the plurality of tiles; and
determining, using the selected neural network model, a change in a temperature of a target tile on an integrated circuit (IC) based on a relationship between a change in temperature relative to a power applied to the IC and a tile-based power map of the IC, the neural network model being selected so the target tile occupies the center tile of the template corresponding to the neural network model.