Search NASA⌕ Search

Engineering topics

Gafur, Jamil

Publications and source records attributed to Gafur, Jamil.

6 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Decision Points and Practical Considerations for AI Projects

In this presentation, I will present business-relevant decisions, risks, and considerations for practical implementations of AI projects. I will use energy efficiency and renewable energy AI projects at NREL as examples and case-studies highlighting the journey from concept to implementation. First, I present challenges, questions, and trade-offs related to system inputs: the data. Next, I will examine issues with system behavior and trust, presenting examples, risks, and mitigation strategies. Finally, I will discuss challenges to effective widespread deployment of AI systems including energy, compute, and time requirements.

AI↗
DOE OSTI2022

BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset

The BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset adds node-level energy consumption data from watt-meters to the primary sweep of the BUTTER - Empirical Deep Learning Dataset. This dataset contains energy consumption and performance data from 63,527 individual experimental runs spanning 30,582 distinct configurations: 13 datasets, 20 sizes (number of trainable parameters), 8 network "shapes", and 14 depths on both CPU and GPU hardware collected using node-level watt-meters. This dataset reveals the complex relationship between dataset size, network structure, and energy use, and highlights the impact of cache effects. BUTTER-E is intended to be joined with the BUTTER dataset (see "BUTTER - Empirical Deep Learning Dataset on OEDI" resource below) which characterizes the performance of 483k distinct fully connected neural networks but does not include energy measurements.

Array↗
DOE OSTI2019

Extending Beryllium Strength Calibration To High Strain-rate Regimes

Current calibration of the PTW strength model for beryllium is accomplished using fits to Hopkinson bar experiments. Unfortunately, Hopkinson bar tests on beryllium only allow for strain-rates of up to $5 * 10^{3} s^{-1}.$ It has been shown previously that, in higher strain-rate experiments such as flyer-plate impact tests, this calibrated strength model causes simulations with beryllium to diverge from the observed response. In this work we use a particle swarm optimization algorithm to arrive at a better calibration of beryllium strength through fits to observed material responses during flyer-plate impact experiments. We show that the PTW strength model for beryllium can be tuned to provide a much better fit to flyer-plate responses. To mitigate the effect of this new calibration on fits to Hopkinson bar data we calibrate to flyer-plate impact data using only parameters within the medium strain-rate regime of the PTW model.

36 MATERIALS SCIENCE↗