Integrated structure selection and parameter optimisation for eng-genes neural models

Patrick Connally, Kang Li, George Irwin

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

The eng-genes concept involves the use of fundamental known system functions as activation functions in a neural model to create a 'grey-box' neural network. One of the main issues in eng-genes modelling is to produce a parsimonious model given a model construction criterion. The challenges are that (1) the eng-genes model in most cases is a heterogenous network consisting of more than one type of nonlinear basis functions, and each basis function may have different set of parameters to be optimised; (2) the number of hidden nodes has to be chosen based on a model selection criterion. This is a mixed integer hard problem and this paper investigates the use of a forward selection algorithm to optimise both the network structure and the parameters of the system-derived activation functions. Results are included from case studies performed on a simulated continuously stirred tank reactor process, and using actual data from a pH neutralisation plant. The resulting eng-genes networks demonstrate superior simulation performance and transparency over a range of network sizes when compared to conventional neural models. (c) 2007 Elsevier B.V. All rights reserved.
Original languageEnglish
Pages (from-to)2964-2977
Number of pages14
JournalNeurocomputing
Volume71
Issue number13-15
DOIs
Publication statusPublished - Aug 2008

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Cognitive Neuroscience

Fingerprint Dive into the research topics of 'Integrated structure selection and parameter optimisation for eng-genes neural models'. Together they form a unique fingerprint.

Cite this