MUMBAI, India, Sept. 28 -- Intellectual Property India has published a patent application (202621097994 A) filed by Lakshmi Narain College Of Technology Excellence on August 13, 2026, for System And Method For Physics-Informed Neural Network-Assisted Tcad Surrogate Modeling For Angstrom-Era Semiconductor Device.
Inventors include Santosh Kumar Jha; Satyarth Tiwari; and Dr. Deepika Sharma.
The application for the patent was published on September 25, 2026, under issue no. 39/2026.
Abstract: ABSTRACT System and Method for Physics-Informed Neural Network-Assisted TCAD Surrogate Modeling for Angstrom-Era Semiconductor Device The present invention is related to a Physics-Informed Neural Network-aided Technology Computer-Aided Design surrogate modelling system and method for designing and optimizing Angstrom era semiconductor devices. This system includes data acquisition engine, semiconductor device repository, processor, memory, Physics-Informed Neural Network kernel, semiconductor physics constraint engine, surrogate modeling engine, uncertainty and validation engine, variability and reliability prediction engine, design space exploration engine, Design Technology Co-optimization engine, and communication interface. The processor collects information on geometry, materials, processing, operating conditions, simulation, and measurements and produces predicted responses for semiconductor devices while ensuring semiconductor physics equations and boundary conditions are enforced. The potential device designs are evaluated based on performance, power, area, fabrication, variability, reliability, yield and prediction accuracy. In the case that the validation condition is not met, then a more accurate TCAD evaluation is conducted. An optimal design is produced only if physics consistency and design constraints are ensured. Fig. 3
Disclaimer: Curated by HT Syndication.
