SCALE RH Machine-learning Based Optimization of Materials for Microelectronics
Category
Published on
Abstract
The purpose of this research is to explore the topic of radiation damage and its effect on material degradation. The effects of radiation damage can be seen in many places, from nuclear systems to space. Current simulation techniques are both expensive and difficult to use. This project aims to utilize machine learning to improve radiation damage simulation. To address these issues, we will first download Geant4 (software toolkit for the simulation of the passage of particles through different materials) through nanoHUB, which is a platform that serves as a collaborative environment where researchers can access a variety of computational and simulation tools. Then, we create a machine learning model that will use data from Geant4 to predict the result of radiation damage. Our results show that we can successfully install Geant4 on nanoHUB. We also notice that after data extraction is complete, we can effectively analyze this data to develop our machine learning model. We expect to train the model and similarly release it publicly to the scientific community, allowing quick and accurate data with respect to radiation damage. Based on this work, we will contribute to the research society by giving access to a tool that provides a quick and accurate radiation damage simulation.
Sponsored by
Purdue First Time Researcher Program
References
- Church, R. (2024, March 2). 16 astonishing facts about space environment effects. Facts.net. https://facts.net/nature/universe/16-astonishing-facts-about-space-environment-effects
- DeepLobe. (2023, September 12). Machine learning for transformers - explained with language translation. https://deeplobe.ai/machine-learning-for-transformers-explained-with-language-translation
Cite this work
Researchers should cite this work as follows: