Curvature-driven smoothing: a learning algorithm for feed-forward networks

The performance of feed-forward neural networks in real applications can be often be improved significantly if use is made of a-priori information. For interpolation problems this prior knowledge frequently includes smoothness requirements on the network mapping, and can be imposed by the addition to the error function of suitable regularization terms. The new error function, however, now depends on the derivatives of the network mapping, and so the standard back-propagation algorithm cannot be applied. In this paper, we derive a computationally efficient learning algorithm, for a feed-forward network of arbitrary topology, which can be used to minimize the new error function. Networks having a single hidden layer, for which the learning algorithm simplifies, are treated as a special case.

Regularization and complexity control in feed-forward networks

In this paper we consider four alternative approaches to complexity control in feed-forward networks based respectively on architecture selection, regularization, early stopping, and training with noise. We show that there are close similarities between these approaches and we argue that, for most practical applications, the technique of regularization should be the method of choice.

Topographic mappings and feed-forward neural networks

This thesis is a study of the generation of topographic mappings - dimension reducing transformations of data that preserve some element of geometric structure - with feed-forward neural networks. As an alternative to established methods, a transformational variant of Sammon's method is proposed, where the projection is effected by a radial basis function neural network. This approach is related to the statistical field of multidimensional scaling, and from that the concept of a 'subjective metric' is defined, which permits the exploitation of additional prior knowledge concerning the data in the mapping process. This then enables the generation of more appropriate feature spaces for the purposes of enhanced visualisation or subsequent classification. A comparison with established methods for feature extraction is given for data taken from the 1992 Research Assessment Exercise for higher educational institutions in the United Kingdom. This is a difficult high-dimensional dataset, and illustrates well the benefit of the new topographic technique. A generalisation of the proposed model is considered for implementation of the classical multidimensional scaling (¸mds}) routine. This is related to Oja's principal subspace neural network, whose learning rule is shown to descend the error surface of the proposed ¸mds model. Some of the technical issues concerning the design and training of topographic neural networks are investigated. It is shown that neural network models can be less sensitive to entrapment in the sub-optimal global minima that badly affect the standard Sammon algorithm, and tend to exhibit good generalisation as a result of implicit weight decay in the training process. It is further argued that for ideal structure retention, the network transformation should be perfectly smooth for all inter-data directions in input space. Finally, there is a critique of optimisation techniques for topographic mappings, and a new training algorithm is proposed. A convergence proof is given, and the method is shown to produce lower-error mappings more rapidly than previous algorithms.

Thermochemical characterisation of straws and high yielding perennial grasses

The research is concerned with thermochemical characterisation of straws and high yielding perennial grasses. Crops selected for this study include wheat straw (Triticum aestivum), rape straw (Brassica napus), reed canary grass (Phalaris arundinacea) and switch grass (Panicum virgatum). Thermogravimetric analysis (TGA) was used to examine the distribution of char and volatiles during pyrolysis up to 900 °C. Utilising multi-heating rate thermogravimetric data, the Friedman iso-conversional kinetic method was used to determine pyrolysis kinetic parameters. Light and medium volatile decomposition products were investigated using pyrolysis–gas chromatography–mass spectrometry (Py–GC–MS) up to 520 °C. The 22 highest yielding identifiable cellulose, hemicellulose and lignin biomass markers were semi-quantified taking into consideration peak areas from GC chromatograms. Notable differences can be seen in butanedioic acid, dimethyl ester (hemicelluloses decomposition products), 2-methoxy-4-vinylphenol (lignin marker) and levoglucosan (intermediate pyrolytic decomposition product of cellulose) content when comparing perennial grasses with straw. From results presented in this study, perennial grasses such as switch grass, have the most attractive properties for fast pyrolysis processing. This is because of the observed high volatile yield content of 82.23%, heating value of 19.64 MJ/kg and the relatively low inorganic content.

Quantification of hydroxycinnamic acids and lignin in perennial forage and energy grasses by Fourier-transform infrared spectroscopy and partial least squares regression

Levels of lignin and hydroxycinnamic acid wall components in three genera of forage grasses (Lolium,Festuca and Dactylis) have been accurately predicted by Fourier-transform infrared spectroscopy using partial least squares models correlated to analytical measurements. Different models were derived that predicted the concentrations of acid detergent lignin, total hydroxycinnamic acids, total ferulate monomers plus dimers, p-coumarate and ferulate dimers in independent spectral test data from methanol extracted samples of perennial forage grass with accuracies of 92.8%, 86.5%, 86.1%, 59.7% and 84.7% respectively, and analysis of model projection scores showed that the models relied generally on spectral features that are known absorptions of these compounds. Acid detergent lignin was predicted in samples of two species of energy grass, (Phalaris arundinacea and Pancium virgatum) with an accuracy of 84.5%.

Measurement of key compositional parameters in two species of energy grass by Fourier transform infrared spectroscopy

Two energy grass species, switch grass, a North American tuft grass, and reed canary grass, a European native, are likely to be important sources of biomass in Western Europe for the production of biorenewable energy. Matching chemical composition to conversion efficiency is a primary goal for improvement programmes and for determining the quality of biomass feed-stocks prior to use and there is a need for methods which allow cost effective characterisation of chemical composition at high rates of sample through-put. In this paper we demonstrate that nitrogen content and alkali index, parameters greatly influencing thermal conversion efficiency, can be accurately predicted in dried samples of these species grown under a range of agronomic conditions by partial least square regression of Fourier transform infrared spectra (R2 values for plots of predicted vs. measured values of 0.938 and 0.937, respectively). We also discuss the prediction of carbon and ash content in these samples and the application of infrared based predictive methods for the breeding improvement of energy grasses.

Prediction of Klason lignin and lignin thermal degradation products by Py-GC/MS in a collection of Lolium and Festuca grasses

A rapid method for the analysis of biomass feedstocks was established to identify the quality of the pyrolysis products likely to impact on bio-oil production. A total of 15 Lolium and Festuca grasses known to exhibit a range of Klason lignin contents were analysed by pyroprobe-GC/MS (Py-GC/MS) to determine the composition of the thermal degradation products of lignin. The identification of key marker compounds which are the derivatives of the three major lignin subunits (G, H, and S) allowed pyroprobe-GC/MS to be statistically correlated to the Klason lignin content of the biomass using the partial least-square method to produce a calibration model. Data from this multivariate modelling procedure was then applied to identify likely "key marker" ions representative of the lignin subunits from the mass spectral data. The combined total abundance of the identified key markers for the lignin subunits exhibited a linear relationship with the Klason lignin content. In addition the effect of alkali metal concentration on optimum pyrolysis characteristics was also examined. Washing of the grass samples removed approximately 70% of the metals and changed the characteristics of the thermal degradation process and products. Overall the data indicate that both the organic and inorganic specification of the biofuel impacts on the pyrolysis process and that pyroprobe-GC/MS is a suitable analytical technique to asses lignin composition. © 2007 Elsevier B.V. All rights reserved.

The characterisation and optimisation of modified herbaceous grasses for identifying pyrolisis-oil quality traits

The primary objective of this work is to relate the biomass fuel quality to fast pyrolysis-oil quality in order to identify key biomass traits which affect pyrolysis-oil stability. During storage the pyrolysis-oil becomes more viscous due to chemical and physical changes, as reactions and volatile losses occur due to aging. The reason for oil instability begins within the pyrolysis reactor during pyrolysis in which the biomass is rapidly heated in the absence of oxygen, producing free radical volatiles which are then quickly condensed to form the oil. The products formed do not reach thermodynamic equilibrium and in tum the products react with each other to try to achieve product stability. The first aim of this research was to develop and validate a rapid screening method for determining biomass lignin content in comparison to traditional, time consuming and hence costly wet chemical methods such as Klason. Lolium and Festuca grasses were selected to validate the screening method, as these grass genotypes exhibit a low range of Klason /Acid Digestible Fibre lignin contents. The screening methodology was based on the relationship between the lignin derived products from pyrolysis and the lignin content as determined by wet chemistry. The second aim of the research was to determine whether metals have an affect on fast pyrolysis products, and if any clear relationships can be deduced to aid research in feedstock selection for fast pyrolysis processing. It was found that alkali metals, particularly Na and K influence the rate and yield of degradation as well the char content. Pre-washing biomass with water can remove 70% of the total metals, and improve the pyrolysis product characteristics by increasing the organic yield, the temperature in which maximum liquid yield occurs and the proportion of higher molecular weight compounds within the pyrolysis-oil. The third aim identified these feedstock traits and relates them to the pyrolysis-oil quality and stability. It was found that the mineral matter was a key determinant on pyrolysis-oil yield compared to the proportion of lignin. However the higher molecular weight compounds present in the pyrolysis-oil are due to the lignin, and can cause instability within the pyrolysis-oil. The final aim was to investigate if energy crops can be enhanced by agronomical practices to produce a biomass quality which is attractive to the biomass conversion community, as well as giving a good yield to the farmers. It was found that the nitrogen/potassium chloride fertiliser treatments enhances Miscanthus qualities, by producing low ash, high volatiles yields with acceptable yields for farmers. The progress of senescence was measured in terms of biomass characteristics and fast pyrolysis product characteristics. The results obtained from this research are in strong agreement with published literature, and provides new information on quality traits for biomass which affects pyrolysis and pyrolysis-oils.

Novel real-time homodyne coherent receiver using a feed-forward based carrier extraction scheme for phase modulated signals

We report a novel real-time homodyne coherent receiver based on a DPSK optical-electrical-optical (OEO) regenerator used to extract a carrier from carrier-less phase modulated signals based on feed-forward based modulation stripping. The performance of this non-DSP based coherent receiver was evaluated for 10.66Gbit/s BPSK signals. Self-homodyne coherent detection and homodyne detection with an injection-locked local oscillator laser was demonstrated. The performance was evaluated by measuring the electrical signal-to-noise (SNR) and recording the eye diagrams. Using injection-locking for the LO improves the performance and enables homodyne detection with optical injection-locking to operate with carrier-less BPSK signals without the need for polarization multiplexed pilot-tones.

Novel synchronous DPSK optical regenerator based on a feed-forward based carrier extraction scheme

We experimentally demonstrate a novel synchronous 10.66Gbit/s DPSK OEO regenerator which uses a feed-forward carrier extraction scheme with an injection-locked laser to synchronize the regenerated signal wavelength to the incoming signal wavelength. After injection-locking, a low-cost DFB laser used at the regenerator exhibited the same linewidth characteristics as the narrow line-width transmitter laser. The phase regeneration properties of the regenerator were evaluated by emulating random Gaussian phase noise applied to the DPSK signal before the regenerator using a phase modulator driven by an arbitrary waveform generator. The overall performance was evaluated in terms of electrical eye-diagrams, BER measurements, and constellation diagrams.

Full-field feed-forward equalizer with adaptive system optimization

We numerically investigate the combination of full-field detection and feed-forward equalizer (FFE) for adaptive chromatic dispersion compensation up to 2160 km in a 10 Gbit/s on-off keyed optical transmission system. The technique, with respect to earlier reports, incorporates several important implementation modules, including the algorithm for adaptive equalization of the gain imbalance between the two receiver chains, compensation of phase misalignment of the asymmetric Mach-Zehnder interferometer, and simplified implementation of field calculation. We also show that in addition to enabling fast adaptation and simplification of field calculation, full-field FFE exhibits enhanced tolerance to the sampling phase misalignment and reduced sampling rate when compared to the full-field implementation using a dispersive transmission line.

An analysis of food demand in Peru:implications for food-feed competition

In their policy proposals on how best to stimulate economic growth, economists have been increasingly emphasizing free markets. It is, however, possible that free-market-led economic growth can lead to increased income inequity which can further increase poverty. One of the more interesting but thus far insufficiently explored mechanisms for the latter is food–feed competition. Using Peruvian Living Standard Survey (PLSS) data for 1985–86 and 1990, the paper examines the demand patterns of households and concludes that the empirical evidence is in agreement with the hypotheses underlying the theory of food–feed competition.