Neural Networks in Dynamic Mechanical Metrology
Develop machine learning systems, especially physics-constrained neural networks, for accurate dynamic mechanical measurements and physics discovery.
This grant is no longer accepting proposals
NRC Research Associateship Programs has archived this opportunity.
Funder: NRC Research Associateship Programs
Due Dates: February 1, 2025 | May 1, 2025 | August 1, 2025 | November 1, 2025
Funding Amounts: $82,764 stipend per year plus $3,000 travel allowance; typical appointment duration 2 years.
Summary: Postdoctoral fellowship to develop physics-constrained neural network systems for accurate dynamic mechanical measurements and physics discovery at NIST.
Key Information: Open to U.S. citizens with a doctoral degree earned within the last 5 years; requires contacting a NIST Research Adviser prior to application.
Description
This fellowship opportunity invites postdoctoral researchers to collaborate with the National Institute of Standards and Technology (NIST) on advancing machine learning applications in dynamic mechanical metrology. The focus is on developing physics-constrained neural network models that can accurately perform deconvolution of sensor outputs to measure dynamic inputs, even when the full system details are unknown. These neural networks accommodate nonlinear behaviors typical in real physical systems and enable transparent models that facilitate physics discovery.
The research involves building neural network representations constrained by physical laws, training these models using known calibration inputs, and applying them to measure arbitrary-waveform unknown inputs with high accuracy. This work integrates machine learning, signal processing, and metrology to improve measurement precision in dynamic mechanical systems.
Key research areas include neural networks, machine learning, artificial intelligence, sensors, signal processing, metrology, measurement, and deconvolution.
