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

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.

A Search for Line Shape and Depth Variations in 51 Pegasi and tau Bootis

Spectroscopic observations of 51 Pegasi and tau Bootis show no periodic changes in the shapes of their line profiles; these results for 51 Peg are in significant conflict with those reported by Gray & Hatzes. Our detection limits are small enough to rule out nonradial pulsations as the cause of the variability in tau Boo, but not in 51 Peg. The absence of line shape changes is consistent with these stars' radial velocity variability arising from planetary mass companions.