An algorithm for optimally fitting a wiener model

Thumbnail Image
Date
2011-01-01
Authors
Beverlin, Lucas
Rollins, Derrick
Vyas, Nisarg
Andre, David
Major Professor
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Authors
Person
Rollins, Derrick K
University Professor Emeritus
Research Projects
Organizational Units
Organizational Unit
Chemical and Biological Engineering

The function of the Department of Chemical and Biological Engineering has been to prepare students for the study and application of chemistry in industry. This focus has included preparation for employment in various industries as well as the development, design, and operation of equipment and processes within industry.Through the CBE Department, Iowa State University is nationally recognized for its initiatives in bioinformatics, biomaterials, bioproducts, metabolic/tissue engineering, multiphase computational fluid dynamics, advanced polymeric materials and nanostructured materials.

History
The Department of Chemical Engineering was founded in 1913 under the Department of Physics and Illuminating Engineering. From 1915 to 1931 it was jointly administered by the Divisions of Industrial Science and Engineering, and from 1931 onward it has been under the Division/College of Engineering. In 1928 it merged with Mining Engineering, and from 1973–1979 it merged with Nuclear Engineering. It became Chemical and Biological Engineering in 2005.

Dates of Existence
1913 - present

Historical Names

  • Department of Chemical Engineering (1913–1928)
  • Department of Chemical and Mining Engineering (1928–1957)
  • Department of Chemical Engineering (1957–1973, 1979–2005)
    • Department of Chemical and Biological Engineering (2005–present)

    Related Units

Journal Issue
Is Version Of
Versions
Series
Abstract

The purpose of this work is to present a new methodology for fitting Wiener networks to datasets with a large number of variables. Wiener networks have the ability to model a wide range of data types, and their structures can yield parameters with phenomenological meaning. There are several challenges to fitting such a model: model stiffness, the nonlinear nature of a Wiener network, possible overfitting, and the large number of parameters inherent with large input sets. This work describes a methodology to overcome these challenges by using several iterative algorithms under supervised learning and fitting subsets of the parameters at a time. This methodology is applied to Wiener networks that are used to predict blood glucose concentrations. The predictions of validation sets from models fit to four subjects using this methodology yielded a higher correlation between observed and predicted observations than other algorithms, including the Gauss-Newton and Levenberg-Marquardt algorithms.

Comments

This article is from Mathematical Problems in Engineering 2011 (2011): article no.570509, doi: 10.1155/2011/570509.

Description
Keywords
Citation
DOI
Copyright
Sat Jan 01 00:00:00 UTC 2011
Collections