Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
individual
Campus
Daytona Beach
Authors' Class Standing
Senior
Lead Presenter's Name
christina bocirnea
Lead Presenter's College
DB College of Arts and Sciences
Faculty Mentor Name
Dr. Siddharth Parida
Abstract
This research enhances a novel finite element physics-informed neural network (FE-PINN) framework in order to optimize efficiency and results. The enhancements include tuning hyperparameters and considering new methodology in constructing the model architecture. This study achieved near convergence of model prediction to actual data and successfully incorporates finite element discretization into a neural network model.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
Yes, Student Internal Grants
FE-PINN Optimization
This research enhances a novel finite element physics-informed neural network (FE-PINN) framework in order to optimize efficiency and results. The enhancements include tuning hyperparameters and considering new methodology in constructing the model architecture. This study achieved near convergence of model prediction to actual data and successfully incorporates finite element discretization into a neural network model.