Prediction- and simulation-error based perceptron training: Solution space analysis and a novel combined training scheme

Previous papers have noted the difficulty in obtaining neural models which are stable under simulation when trained using prediction-error-based methods. Here the differences between series-parallel and parallel identification structures for training neural models are investigated. The effect of the error surface shape on training convergence and simulation performance is analysed using a standard algorithm operating in both training modes. A combined series-parallel/parallel training scheme is proposed, aiming to provide a more effective means of obtaining accurate neural simulation models. Simulation examples show the combined scheme is advantageous in circumstances where the solution space is known or suspected to be complex. (c) 2006 Elsevier B.V. All rights reserved.

Thermogravimetric evaluation of perennial ryegrass (Lolium perenne) for the prediction of in vitro dry matter digestibility

Thermogravimetry (TG) can be used for assessing the compositional differences in grasses that relate to dry matter digestibility (DMD) determined by pepsin-cellulase assay. This investigation developed regression models for predicting DMD of herbage grass during one growing season using TG results. The calibration samples were obtained from a field trial of eight cultivars and two breeding lines. The harvested materials from five cuts were analysed by TG to identify differences in the combustion patterns within the range of 30-600 degrees C. The discrete results including weight loss, peak height, area, temperature, widths and residue of three decomposition peaks were regressed against the measured DMD values of the calibration samples. Similarly, continuous weight loss results of the same samples were also utilised to generate DMD models. The r(2) for validation of the discrete and the best continuous models were 0.90 and 0.95, respectively, and the two calibrations were validated using independent samples from 24 plots from a trial carried out in 2004. The standard error for prediction of the 24 samples by the discrete model (4.14%) was higher than that by the continuous model (2.98%). This study has shown that DMD of grass could be predicted from the TG results. The benefit of thermal analysis is the ability to detect and show changes in composition of cell wall fractions of grasses during different cuts in a year.

Prediction of ionic liquid properties. I. Volumetric properties as a function of temperature at 0.1 MPa

The prediction of molar volumes and densities of several ionic liquids has been achieved using a group contribution model as a function of temperature between (273 and 423) K at atmospheric pressure. It was observed that the calculation of molar volumes or densities could be performed using the "ideal" behavior of the molar volumes of mixtures of ionic liquids. This model is based on the observations of Canongia Lopes et al. (J. Phys. Chem. B 2005, 109, 3519-3525) which showed that this ideal behavior is independent of the temperature and allows the molar volume of a given ionic liquid to be calculated by the sum of the effective molar volume of the component ions. Using this assumption, the effective molar volumes of ions constituting more than 220 different ionic liquids were calculated as a function of the temperature at 0.1 MPa using more than 2150 data points. These calculated results were used to build up a group contribution model for the calculation of ionic liquid molar volumes and densities with an estimated repeatability and uncertainty of 0.36% and 0.48%, respectively. The impact of impurities (water and halide content) in ionic liquids as well as the method of determination were also analyzed and quantified to estimate the overall uncertainty. © 2008 American Chemical Society.