CMU MEMS Laboratory Publication Abstract

 

in Technical Proceedings of Eighth International Conference on Modeling and Simulation of Microsystems, vol. 1 (MSM), pp. 616-619, May 8-12, 2005, Anaheim, CA.
Microfluidic Injector Models Based on Neural Networks
R. Magargle, J. Hoburg and T. Mukherjee
ABSTRACT:
A functional modeling technique is developed for components of a microfluidic system and applied to three common injector topologies. This technique uses sparse numerical simulation to train a neural network to provide compact, explicit, and accurate component models. The resulting models are compatible with analytic system simulation environments, making complex design synthesis and optimization feasible, unlike standard techniques using computationally expensive numerical simulation. The neural network models are accurate to numerical simulation with mean squared errors less than 10-4. In explicit form, the neural network models are very fast taking less than 1s per evaluation.
© 2005 Computational Publications. Abstracting is permitted with credit to the source. Other copying, reprint or repoduction requests should be addressed to: Copyrights Manager, Computational Publications, Copyright Office, 899 Rue Jean de Gingins, 01220 Divonne les Bains, France. Computational Publications is a subsuduary of the Applied Computational Researh Society, a non-profit organization
Full paper not available from outside CMU


This page was generated in 0.006897 seconds at 04:23:03 am EST on 20 Nov 2017.

overview | projects | people | publications | intranet | resources         © 1998-2009  Carnegie Mellon