First year qualifying report: neural networks for extracting wind vectors from satellite scatterometer data

The ERS-1 Satellite was launched in July 1991 by the European Space Agency into a polar orbit at about km800, carrying a C-band scatterometer. A scatterometer measures the amount of radar back scatter generated by small ripples on the ocean surface induced by instantaneous local winds. Operational methods that extract wind vectors from satellite scatterometer data are based on the local inversion of a forward model, mapping scatterometer observations to wind vectors, by the minimisation of a cost function in the scatterometer measurement space.par This report uses mixture density networks, a principled method for modelling conditional probability density functions, to model the joint probability distribution of the wind vectors given the satellite scatterometer measurements in a single cell (the `inverse' problem). The complexity of the mapping and the structure of the conditional probability density function are investigated by varying the number of units in the hidden layer of the multi-layer perceptron and the number of kernels in the Gaussian mixture model of the mixture density network respectively. The optimal model for networks trained per trace has twenty hidden units and four kernels. Further investigation shows that models trained with incidence angle as an input have results comparable to those models trained by trace. A hybrid mixture density network that incorporates geophysical knowledge of the problem confirms other results that the conditional probability distribution is dominantly bimodal.par The wind retrieval results improve on previous work at Aston, but do not match other neural network techniques that use spatial information in the inputs, which is to be expected given the ambiguity of the inverse problem. Current work uses the local inverse model for autonomous ambiguity removal in a principled Bayesian framework. Future directions in which these models may be improved are given.

Dynamics of learning with restricted training sets

The dynamics of supervised learning in layered neural networks were studied in the regime where the size of the training set is proportional to the number of inputs. The evolution of macroscopic observables, including the two relevant performance measures can be predicted by using the dynamical replica theory. Three approximation schemes aimed at eliminating the need to solve a functional saddle-point equation at each time step have been derived.

Robust automatic mapping algorithms in a network monitoring scenario

Automatically generating maps of a measured variable of interest can be problematic. In this work we focus on the monitoring network context where observations are collected and reported by a network of sensors, and are then transformed into interpolated maps for use in decision making. Using traditional geostatistical methods, estimating the covariance structure of data collected in an emergency situation can be difficult. Variogram determination, whether by method-of-moment estimators or by maximum likelihood, is very sensitive to extreme values. Even when a monitoring network is in a routine mode of operation, sensors can sporadically malfunction and report extreme values. If this extreme data destabilises the model, causing the covariance structure of the observed data to be incorrectly estimated, the generated maps will be of little value, and the uncertainty estimates in particular will be misleading. Marchant and Lark [2007] propose a REML estimator for the covariance, which is shown to work on small data sets with a manual selection of the damping parameter in the robust likelihood. We show how this can be extended to allow treatment of large data sets together with an automated approach to all parameter estimation. The projected process kriging framework of Ingram et al. [2007] is extended to allow the use of robust likelihood functions, including the two component Gaussian and the Huber function. We show how our algorithm is further refined to reduce the computational complexity while at the same time minimising any loss of information. To show the benefits of this method, we use data collected from radiation monitoring networks across Europe. We compare our results to those obtained from traditional kriging methodologies and include comparisons with Box-Cox transformations of the data. We discuss the issue of whether to treat or ignore extreme values, making the distinction between the robust methods which ignore outliers and transformation methods which treat them as part of the (transformed) process. Using a case study, based on an extreme radiological events over a large area, we show how radiation data collected from monitoring networks can be analysed automatically and then used to generate reliable maps to inform decision making. We show the limitations of the methods and discuss potential extensions to remedy these.

Use of computational fluid dynamics in the design of bubble column reactors

Investigations into the modelling techniques that depict the transport of discrete phases (gas bubbles or solid particles) and model biochemical reactions in a bubble column reactor are discussed here. The mixture model was used to calculate gas-liquid, solid-liquid and gasliquid-solid interactions. Multiphase flow is a difficult phenomenon to capture, particularly in bubble columns where the major driving force is caused by the injection of gas bubbles. The gas bubbles cause a large density difference to occur that results in transient multi-dimensional fluid motion. Standard design procedures do not account for the transient motion, due to the simplifying assumptions of steady plug flow. Computational fluid dynamics (CFD) can assist in expanding the understanding of complex flows in bubble columns by characterising the flow phenomena for many geometrical configurations. Therefore, CFD has a role in the education of chemical and biochemical engineers, providing the examples of flow phenomena that many engineers may not experience, even through experimentation. The performance of the mixture model was investigated for three domains (plane, rectangular and cylindrical) and three flow models (laminar, k-e turbulence and the Reynolds stresses). mThis investigation raised many questions about how gas-liquid interactions are captured numerically. To answer some of these questions the analogy between thermal convection in a cavity and gas-liquid flow in bubble columns was invoked. This involved modelling the buoyant motion of air in a narrow cavity for a number of turbulence schemes. The difference in density was caused by a temperature gradient that acted across the width of the cavity. Multiple vortices were obtained when the Reynolds stresses were utilised with the addition of a basic flow profile after each time step. To implement the three-phase models an alternative mixture model was developed and compared against a commercially available mixture model for three turbulence schemes. The scheme where just the Reynolds stresses model was employed, predicted the transient motion of the fluids quite well for both mixture models. Solid-liquid and then alternative formulations of gas-liquid-solid model were compared against one another. The alternative form of the mixture model was found to perform particularly well for both gas and solid phase transport when calculating two and three-phase flow. The improvement in the solutions obtained was a result of the inclusion of the Reynolds stresses model and differences in the mixture models employed. The differences between the alternative mixture models were found in the volume fraction equation (flux and deviatoric stress tensor terms) and the viscosity formulation for the mixture phase.

A constructive learning algorithm based on back-propagation

There are been a resurgence of interest in the neural networks field in recent years, provoked in part by the discovery of the properties of multi-layer networks. This interest has in turn raised questions about the possibility of making neural network behaviour more adaptive by automating some of the processes involved. Prior to these particular questions, the process of determining the parameters and network architecture required to solve a given problem had been a time consuming activity. A number of researchers have attempted to address these issues by automating these processes, concentrating in particular on the dynamic selection of an appropriate network architecture.The work presented here specifically explores the area of automatic architecture selection; it focuses upon the design and implementation of a dynamic algorithm based on the Back-Propagation learning algorithm. The algorithm constructs a single hidden layer as the learning process proceeds using individual pattern error as the basis of unit insertion. This algorithm is applied to several problems of differing type and complexity and is found to produce near minimal architectures that are shown to have a high level of generalisation ability.

Aspects of design for a spare parts provisioning system

This thesis describes an investigation by the author into the spares operation of compare BroomWade Ltd. Whilst the complete system, including the warehousing and distribution functions, was investigated, the thesis concentrates on the provisioning aspect of the spares supply problem. Analysis of the historical data showed the presence of significant fluctuations in all the measures of system performance. Two Industrial Dynamics simulation models were developed to study this phenomena. The models showed that any fluctuation in end customer demand would be amplified as it passed through the distributor and warehouse stock control systems. The evidence from the historical data available supported this view of the system's operation. The models were utilised to determine which parts of the total system could be expected to exert a critical influence on its performance. The lead time parameters of the supply sector were found to be critical and further study showed that the manner in which the lead time changed with work in progress levels was also an important factor. The problem therefore resolved into the design of a spares manufacturing system. Which exhibited the appropriate dynamic performance characteristics. The gross level of entity presentation, inherent in the Industrial Dynamics methodology, was found to limit the value of these models in the development of detail design proposals. Accordingly, an interacting job shop simulation package was developed to allow detailed evaluation of organisational factors on the performance characteristics of a manufacturing system. The package was used to develop a design for a pilot spares production unit. The need for a manufacturing system to perform successfully under conditions of fluctuating demand is not limited to the spares field. Thus, although the spares exercise provides an example of the approach, the concepts and techniques developed can be considered to have broad application throughout batch manufacturing industry.

Investigating viscous fluid flow in an internal mixer using computational fluid dynamics

This thesis presents an effective methodology for the generation of a simulation which can be used to increase the understanding of viscous fluid processing equipment and aid in their development, design and optimisation. The Hampden RAPRA Torque Rheometer internal batch twin rotor mixer has been simulated with a view to establishing model accuracies, limitations, practicalities and uses. As this research progressed, via the analyses several 'snap-shot' analysis of several rotor configurations using the commercial code Polyflow, it was evident that the model was of some worth and its predictions are in good agreement with the validation experiments, however, several major restrictions were identified. These included poor element form, high man-hour requirements for the construction of each geometry and the absence of the transient term in these models. All, or at least some, of these limitations apply to the numerous attempts to model internal mixes by other researchers and it was clear that there was no generally accepted methodology to provide a practical three-dimensional model which has been adequately validated. This research, unlike others, presents a full complex three-dimensional, transient, non-isothermal, generalised non-Newtonian simulation with wall slip which overcomes these limitations using unmatched ridding and sliding mesh technology adapted from CFX codes. This method yields good element form and, since only one geometry has to be constructed to represent the entire rotor cycle, is extremely beneficial for detailed flow field analysis when used in conjunction with user defined programmes and automatic geometry parameterisation (AGP), and improves accuracy for investigating equipment design and operation conditions. Model validation has been identified as an area which has been neglected by other researchers in this field, especially for time dependent geometries, and has been rigorously pursued in terms of qualitative and quantitative velocity vector analysis of the isothermal, full fill mixing of generalised non-Newtonian fluids, as well as torque comparison, with a relatively high degree of success. This indicates that CFD models of this type can be accurate and perhaps have not been validated to this extent previously because of the inherent difficulties arising from most real processes.

A network coding based interference cancelation scheme for wireless ad hoc networks

The performance of wireless networks is limited by multiple access interference (MAI) in the traditional communication approach where the interfered signals of the concurrent transmissions are treated as noise. In this paper, we treat the interfered signals from a new perspective on the basis of additive electromagnetic (EM) waves and propose a network coding based interference cancelation (NCIC) scheme. In the proposed scheme, adjacent nodes can transmit simultaneously with careful scheduling; therefore, network performance will not be limited by the MAI. Additionally we design a space segmentation method for general wireless ad hoc networks, which organizes network into clusters with regular shapes (e.g., square and hexagon) to reduce the number of relay nodes. The segmentation methodworks with the scheduling scheme and can help achieve better scalability and reduced complexity. We derive accurate analytic models for the probability of connectivity between two adjacent cluster heads which is important for successful information relay. We proved that with the proposed NCIC scheme, the transmission efficiency can be improved by at least 50% for general wireless networks as compared to the traditional interference avoidance schemes. Numeric results also show the space segmentation is feasible and effective. Finally we propose and discuss a method to implement the NCIC scheme in a practical orthogonal frequency division multiplexing (OFDM) communications networks. Copyright © 2009 John Wiley & Sons, Ltd.

Network lifetime maximization with cross-layer design in wireless sensor networks

This paper investigates a cross-layer design approach for minimizing energy consumption and maximizing network lifetime (NL) of a multiple-source and single-sink (MSSS) WSN with energy constraints. The optimization problem for MSSS WSN can be formulated as a mixed integer convex optimization problem with the adoption of time division multiple access (TDMA) in medium access control (MAC) layer, and it becomes a convex problem by relaxing the integer constraint on time slots. Impacts of data rate, link access and routing are jointly taken into account in the optimization problem formulation. Both linear and planar network topologies are considered for NL maximization (NLM). With linear MSSS and planar single-source and single-sink (SSSS) topologies, we successfully use Karush-Kuhn-Tucker (KKT) optimality conditions to derive analytical expressions of the optimal NL when all nodes are exhausted simultaneously. The problem for planar MSSS topology is more complicated, and a decomposition and combination (D&C) approach is proposed to compute suboptimal solutions. An analytical expression of the suboptimal NL is derived for a small scale planar network. To deal with larger scale planar network, an iterative algorithm is proposed for the D&C approach. Numerical results show that the upper-bounds of the network lifetime obtained by our proposed optimization models are tight. Important insights into the NL and benefits of cross-layer design for WSN NLM are obtained.

Investigation of EDFA power transients in circuit-switched and packet-switched optical networks

Erbium-doped fibre amplifiers (EDFA’s) are a key technology for the design of all optical communication systems and networks. The superiority of EDFAs lies in their negligible intermodulation distortion across high speed multichannel signals, low intrinsic losses, slow gain dynamics, and gain in a wide range of optical wavelengths. Due to long lifetime in excited states, EDFAs do not oppose the effect of cross-gain saturation. The time characteristics of the gain saturation and recovery effects are between a few hundred microseconds and 10 milliseconds. However, in wavelength division multiplexed (WDM) optical networks with EDFAs, the number of channels traversing an EDFA can change due to the faulty link of the network or the system reconfiguration. It has been found that, due to the variation in channel number in the EDFAs chain, the output system powers of surviving channels can change in a very short time. Thus, the power transient is one of the problems deteriorating system performance. In this thesis, the transient phenomenon in wavelength routed WDM optical networks with EDFA chains was investigated. The task was performed using different input signal powers for circuit switched networks. A simulator for the EDFA gain dynamicmodel was developed to compute the magnitude and speed of the power transients in the non-self-saturated EDFA both single and chained. The dynamic model of the self-saturated EDFAs chain and its simulator were also developed to compute the magnitude and speed of the power transients and the Optical signal-to-noise ratio (OSNR). We found that the OSNR transient magnitude and speed are a function of both the output power transient and the number of EDFAs in the chain. The OSNR value predicts the level of the quality of service in the related network. It was found that the power transients for both self-saturated and non-self-saturated EDFAs are close in magnitude in the case of gain saturated EDFAs networks. Moreover, the cross-gain saturation also degrades the performance of the packet switching networks due to varying traffic characteristics. The magnitude and the speed of output power transients increase along the EDFAs chain. An investigation was done on the asynchronous transfer mode (ATM) or the WDM Internet protocol (WDM-IP) traffic networks using different traffic patterns based on the Pareto and Poisson distribution. The simulator is used to examine the amount and speed of the power transients in Pareto and Poisson distributed traffic at different bit rates, with specific focus on 2.5 Gb/s. It was found from numerical and statistical analysis that the power swing increases if the time interval of theburst-ON/burst-OFF is long in the packet bursts. This is because the gain dynamics is fast during strong signal pulse or with long duration pulses, which is due to the stimulatedemission avalanche depletion of the excited ions. Thus, an increase in output power levelcould lead to error burst which affects the system performance.

Market-driven strategies and the design of flexible production systems: evidence from the electronics industry

Links the concept of market-driven business strategies with the design of production systems. It draws upon the case of a firm which, during the last decade, changed its strategy from being “technology led” to “market driven”. The research, based on interdisciplinary fieldwork involving long-term participant observation, investigated the factors which contribute to the successful design and implementation of flexible production systems in electronics assembly. These investigations were conducted in collaboration with a major computer manufacturer, with other electronics firms being studied for comparison. The research identified a number of strategies and actions seen as crucial to the development of efficient flexible production systems, namely: effective integration of subsystems, development of appropriate controls and performance measures, compatibility between production system design and organization structure, and the development of a climate conducive to organizational change. Overall, the analysis suggests that in the electronics industry there exists an extremely high degree of environmental complexity and turbulence. This serves to shape the strategic, technical and social structures that are developed to match this complexity, examples of which are niche marketing, flexible manufacturing and employee harmonization.

Modular system design and control for flexible assembly

Flexible Assembly Systems (FASs) are normally associated with the automatic, or robotic, assembly of products, supported by automated material handling systems. However, manual assembly operations are still prevalent within many industries, where the complexity and variety of products prohibit the development of suitable automated assembly equipment. This article presents a generic model for incorporating flexibility into the design and control of assembly operations concerned with high variety/low volume manufacture, drawing on the principles for Flexible Manufacturing Systems (FMS) and Just-in-Time (JIT) delivery. It is based on work being undertaken in an electronics company where the assembly operations have been overhauled and restructured in response to a need for greater flexibility, shorter cycle times and reduced inventory levels. The principles employed are in themselves not original. However, the way they have been combined and tailored has created a total manufacturing control system which represents a new concept for responding to demands placed on market driven firms operating in an uncertain environment.

Was it good for them? Reflections on the use of MENTOR and changes in course design using general performance measures

This paper considers the use of general performance measures in evaluating specific planning and design decisions in higher education and reflects on the students' learning process. Specifically, it concerns the use of the MENTOR multimedia computer aided learning package for helping students learn about OR as part of a general business degree. It includes the transfer of responsibility for a learning module to a new staff member and a change from a single tutor to a system involving multiple tutors. Student satisfaction measures, learning outcome measures and MENTOR usage patterns are examined in monitoring the effects of the changes in course delivery. The results raise some questions about the effectiveness of general performance measures in supporting specific decisions relating to course design and planning.