Bayesian hybrid algorithms and models : implementation and associated issues

This thesis addresses computational challenges arising from Bayesian analysis of complex real-world problems. Many of the models and algorithms designed for such analysis are ‘hybrid’ in nature, in that they are a composition of components for which their individual properties may be easily described but the performance of the model or algorithm as a whole is less well understood. The aim of this research project is to after a better understanding of the performance of hybrid models and algorithms. The goal of this thesis is to analyse the computational aspects of hybrid models and hybrid algorithms in the Bayesian context. The first objective of the research focuses on computational aspects of hybrid models, notably a continuous finite mixture of t-distributions. In the mixture model, an inference of interest is the number of components, as this may relate to both the quality of model fit to data and the computational workload. The analysis of t-mixtures using Markov chain Monte Carlo (MCMC) is described and the model is compared to the Normal case based on the goodness of fit. Through simulation studies, it is demonstrated that the t-mixture model can be more flexible and more parsimonious in terms of number of components, particularly for skewed and heavytailed data. The study also reveals important computational issues associated with the use of t-mixtures, which have not been adequately considered in the literature. The second objective of the research focuses on computational aspects of hybrid algorithms for Bayesian analysis. Two approaches will be considered: a formal comparison of the performance of a range of hybrid algorithms and a theoretical investigation of the performance of one of these algorithms in high dimensions. For the first approach, the delayed rejection algorithm, the pinball sampler, the Metropolis adjusted Langevin algorithm, and the hybrid version of the population Monte Carlo (PMC) algorithm are selected as a set of examples of hybrid algorithms. Statistical literature shows how statistical efficiency is often the only criteria for an efficient algorithm. In this thesis the algorithms are also considered and compared from a more practical perspective. This extends to the study of how individual algorithms contribute to the overall efficiency of hybrid algorithms, and highlights weaknesses that may be introduced by the combination process of these components in a single algorithm. The second approach to considering computational aspects of hybrid algorithms involves an investigation of the performance of the PMC in high dimensions. It is well known that as a model becomes more complex, computation may become increasingly difficult in real time. In particular the importance sampling based algorithms, including the PMC, are known to be unstable in high dimensions. This thesis examines the PMC algorithm in a simplified setting, a single step of the general sampling, and explores a fundamental problem that occurs in applying importance sampling to a high-dimensional problem. The precision of the computed estimate from the simplified setting is measured by the asymptotic variance of the estimate under conditions on the importance function. Additionally, the exponential growth of the asymptotic variance with the dimension is demonstrated and we illustrates that the optimal covariance matrix for the importance function can be estimated in a special case.

Temperature and electric field dependent mobility in poly„3-hexylthiophene diodes

Current-voltage (I-V) curves of Poly(3-hexyl-thiophene) (P3HT) diodes have been collected to investigate the polymer hole-dominated charge transport. At room temperature and at low electric fields the I-V characteristic is purely Ohmic whereas at medium-high electric fields, experimental data shows that the hole transport is Trap Dominated - Space Charge Limited Current (TD-SCLC). In this regime, it is possible to extract the I-V characteristic of the P3HT/Al junction showing the ideal Schottky diode behaviour over five orders of magnitude. At high-applied electric fields, holes’ transport is found to be in the trap free SCLC regime. We have measured and modelled in this regime the holes’ mobility to evaluate its dependence from the electric field applied and the temperature of the device.

Flexible high-power multi DC-DC converters for train systems

This thesis reports on the investigations, simulations and analyses of novel power electronics topologies and control strategies. The research is financed by an Australian Research Council (ARC) Linkage (07-09) grant. Therefore, in addition to developing original research and contributing to the available knowledge of power electronics, it also contributes to the design of a DC-DC converter for specific application to the auxiliary power supply in electric trains. Specifically, in this regard, it contributes to the design of a 7.5 kW DC-DC converter for the industrial partner (Schaffler and Associates Ltd) who supported this project. As the thesis is formatted as a ‘thesis by publication’, the contents are organized around published papers. The research has resulted in eleven papers, including seven peer reviewed and published conference papers, one published journal paper, two journal papers accepted for publication and one submitted journal paper (provisionally accepted subject to few changes). In this research, several novel DC-DC converter topologies are introduced, analysed, and tested. The similarity of all of the topologies devised lies in their ‘current circulating’ switching state, which allows them to store some energy in the inductor, as extra inductor current. The stored energy may be applied to enhance the performance of the converter in the occurrence of load current or input voltage disturbances. In addition, when there is an alternating load current, the ability to store energy allows the converter to perform satisfactorily despite frequently and highly varying load current. In this research, the capability of current storage has been utilised to design topologies for specific applications, and the enhancement of the performance of the considered applications has been illustrated. The simplest DC-DC converter topology, which has a ‘current circulating’ switching state, is the Positive Buck-Boost (PBB) converter (also known as the non-inverting Buck-Boost converter). Usually, the topology of the PBB converter is operating as a Buck or a Boost converter in applications with widely varying input voltage or output reference voltage. For example, in electric railways (the application of our industrial partner), the overhead line voltage alternates from 1000VDC to 500VDC and the required regulated voltage is 600VDC. In the course of this research, our industrial partner (Schaffler and Associates Ltd) industrialized a PBB converter–the ‘Mudo converter’–operating at 7.5 kW. Programming the onboard DSP and testing the PBB converter in experimental and nominal power and voltage was part of this research program. In the earlier stages of this research, the advantages and drawbacks of utilization of the ‘current circulating’ switching state in the positive Buck-Boost converter were investigated. In brief, the advantages were found to be robustness against input voltage and current load disturbances, and the drawback was extra conduction and switching loss. Although the robustness against disturbances is desirable for many applications, the price of energy loss must be minimized to attract attention to the utilization of the PBB converter. In further stages of this research, two novel control strategies for different applications were devised to minimise the extra energy loss while the advantages of the positive Buck-Boost converter were fully utilized. The first strategy is Smart Load Controller (SLC) for applications with pre-knowledge or predictability of input voltage and/or load current disturbances. A convenient example of these applications is electric/hybrid cars where a master controller commands all changes in loads and voltage sources. Therefore, the master controller has a pre-knowledge of the load and input voltage disturbances so it can apply the SLC strategy to utilize robustness of the PBB converter. Another strategy aiming to minimise energy loss and maximise the robustness in the face of disturbance is developed to cover applications with unexpected disturbances. This strategy is named Dynamic Hysteresis Band (DHB), and is used to manipulate the hysteresis band height after occurrence of disturbance to reduce dynamics of the output voltage. When no disturbance has occurred, the PBB converter works with minimum inductor current and minimum energy loss. New topologies based on the PBB converter have been introduced to address input voltage disturbances for different onboard applications. The research shows that the performance of applications of symmetrical/asymmetrical multi-level diode-clamped inverters, DC-networks, and linear-assisted RF amplifiers may be enhanced by the utilization of topologies based on the PBB converter. Multi-level diode-clamped inverters have the problem of DC-link voltage balancing when the power factor of their load closes to unity. This research has shown that this problem may be solved with a suitable multi-output DC-DC converter supplying DClink capacitors. Furthermore, the multi-level diode-clamped inverters supplied with asymmetrical DC-link voltages may improve the quality of load voltage and reduce the level of Electromagnetic Interference (EMI). Mathematical analyses and experiments on supplying symmetrical and asymmetrical multi-level inverters by specifically designed multi-output DC-DC converters have been reported in two journal papers. Another application in which the system performance can be improved by utilization of the ‘current circulating’ switching state is linear-assisted RF amplifiers in communicational receivers. The concept of ‘linear-assisted’ is to divide the signal into two frequency domains: low frequency, which should be amplified by a switching circuit; and the high frequency domain, which should be amplified by a linear amplifier. The objective is to minimize the overall power loss. This research suggests using the current storage capacity of a PBB based converter to increase its bandwidth, and to increase the domain of the switching converter. The PBB converter addresses the industrial demand for a DC-DC converter for the application of auxiliary power supply of a typical electric train. However, after testing the industrial prototype of the PBB converter, there were some voltage and current spikes because of switching. To attenuate this problem without significantly increasing the switching loss, the idea of Active Gate Signalling (AGS) is presented. AGS suggests a smart gate driver that selectively controls the switching process to reduce voltage/current spikes, without unacceptable reduction in the efficiency of switching.

Audio visual automatic speech recognition in vehicles

Acoustically, car cabins are extremely noisy and as a consequence, existing audio-only speech recognition systems, for voice-based control of vehicle functions such as the GPS based navigator, perform poorly. Audio-only speech recognition systems fail to make use of the visual modality of speech (eg: lip movements). As the visual modality is immune to acoustic noise, utilising this visual information in conjunction with an audio only speech recognition system has the potential to improve the accuracy of the system. The field of recognising speech using both auditory and visual inputs is known as Audio Visual Speech Recognition (AVSR). Continuous research in AVASR field has been ongoing for the past twenty-five years with notable progress being made. However, the practical deployment of AVASR systems for use in a variety of real-world applications has not yet emerged. The main reason is due to most research to date neglecting to address variabilities in the visual domain such as illumination and viewpoint in the design of the visual front-end of the AVSR system. In this paper we present an AVASR system in a real-world car environment using the AVICAR database [1], which is publicly available in-car database and we show that the use of visual speech conjunction with the audio modality is a better approach to improve the robustness and effectiveness of voice-only recognition systems in car cabin environments.

A hybrid genetic algorithm for the optimal constrained web service selection problem in web service composition

Web service composition is an important problem in web service based systems. It is about how to build a new value-added web service using existing web services. A web service may have many implementations, all of which have the same functionality, but may have different QoS values. Thus, a significant research problem in web service composition is how to select a web service implementation for each of the web services such that the composite web service gives the best overall performance. This is so-called optimal web service selection problem. There may be mutual constraints between some web service implementations. Sometimes when an implementation is selected for one web service, a particular implementation for another web service must be selected. This is so called dependency constraint. Sometimes when an implementation for one web service is selected, a set of implementations for another web service must be excluded in the web service composition. This is so called conflict constraint. Thus, the optimal web service selection is a typical constrained ombinatorial optimization problem from the computational point of view. This paper proposes a new hybrid genetic algorithm for the optimal web service selection problem. The hybrid genetic algorithm has been implemented and evaluated. The evaluation results have shown that the hybrid genetic algorithm outperforms other two existing genetic algorithms when the number of web services and the number of constraints are large.

A Novel Approach to Haptic Tele-operation of Aerial Robot Vehicles

We present a novel, simple and effective approach for tele-operation of aerial robotic vehicles with haptic feedback. Such feedback provides the remote pilot with an intuitive feel of the robot’s state and perceived local environment that will ensure simple and safe operation in cluttered 3D environments common in inspection and surveillance tasks. Our approach is based on energetic considerations and uses the concepts of network theory and port-Hamiltonian systems. We provide a general framework for addressing problems such as mapping the limited stroke of a ‘master’ joystick to the infinite stroke of a ‘slave’ vehicle, while preserving passivity of the closed-loop system in the face of potential time delays in communications links and limited sensor data

A new framework for force feedback teleoperation of robotic vehicles based on optical flow

This paper proposes the use of optical flow from a moving robot to provide force feedback to an operator's joystick to facilitate collision free teleoperation. Optic flow is measured by wide angle cameras on board the vehicle and used to generate a virtual environmental force that is reflected to the user through the joystick, as well as feeding back into the control of the vehicle. The coupling between optical flow (velocity) and force is modelled as an impedance - in this case an optical impedance. We show that the proposed control is dissipative and prevents the vehicle colliding with the environment as well as providing the operator with a natural feel for the remote environment. The paper focuses on applications to aerial robotics vehicles, however, the ideas apply directly to other force actuated vehicles such as submersibles or space vehicles, and the authors believe the approach has potential for control of terrestrial vehicles and even teleoperation of manipulators. Experimental results are provided for a simulated aerial robot in a virtual environment controlled by a haptic joystick.

Reactive navigation and opportunistic localization for autonomous underground mining vehicles

This paper describes an autonomous navigation system for a large underground mining vehicle. The control architecture is based on a robust reactive wall-following behaviour. To make it purposeful we provide driving hints derived from an approximate nodal-map. For most of the time, the vehicle is driven with weak localization (odometry). This need only be improved at intersections where decisions must be made – a technique we refer to as opportunistic localization. The paper briefly reviews absolute and relative navigation strategies, and describes an implementation of a reactive navigation system on a 30 tonne Load-Haul-Dump truck. This truck has achieved full-speed autonomous operation at an artificial test mine, and subsequently, at a operational underground mine.

UAS Mission Path Planning System, (MPPS) using hybrid-game coupled to multi-objective optimizer

This paper presents the application of advanced optimization techniques to unmanned aerial system mission path planning system (MPPS) using multi-objective evolutionary algorithms (MOEAs). Two types of multi-objective optimizers are compared; the MOEA nondominated sorting genetic algorithm II and a hybrid-game strategy are implemented to produce a set of optimal collision-free trajectories in a three-dimensional environment. The resulting trajectories on a three-dimensional terrain are collision-free and are represented by using Bézier spline curves from start position to target and then target to start position or different positions with altitude constraints. The efficiency of the two optimization methods is compared in terms of computational cost and design quality. Numerical results show the benefits of adding a hybrid-game strategy to a MOEA and for a MPPS.

Active flow control bump design using hybrid Nash-game coupled to evolutionary algorithms

In this paper, the optimal design of an active flow control device; Shock Control Bump (SCB) on suction and pressure sides of transonic aerofoil to reduce transonic total drag is investigated. Two optimisation test cases are conducted using different advanced Evolutionary Algorithms (EAs); the first optimiser is the Hierarchical Asynchronous Parallel Evolutionary Algorithm (HAPMOEA) based on canonical Evolutionary Strategies (ES). The second optimiser is the HAPMOEA is hybridised with one of well-known Game Strategies; Nash-Game. Numerical results show that SCB significantly reduces the drag by 30% when compared to the baseline design. In addition, the use of a Nash-Game strategy as a pre-conditioner of global control saves computational cost up to 90% when compared to the first optimiser HAPMOEA.

On the optimal capacity expansion of electric power systems

The analysis of investment in the electric power has been the subject of intensive research for many years. The efficient generation and distribution of electrical energy is a difficult task involving the operation of a complex network of facilities, often located over very large geographical regions. Electric power utilities have made use of an enormous range of mathematical models. Some models address time spans which last for a fraction of a second, such as those that deal with lightning strikes on transmission lines while at the other end of the scale there are models which address time horizons consisting of ten or twenty years; these usually involve long range planning issues. This thesis addresses the optimal long term capacity expansion of an interconnected power system. The aim of this study has been to derive a new, long term planning model which recognises the regional differences which exist for energy demand and which are present in the construction and operation of power plant and transmission line equipment. Perhaps the most innovative feature of the new model is the direct inclusion of regional energy demand curves in the nonlinear form. This results in a nonlinear capacity expansion model. After review of the relevant literature, the thesis first develops a model for the optimal operation of a power grid. This model directly incorporates regional demand curves. The model is a nonlinear programming problem containing both integer and continuous variables. A solution algorithm is developed which is based upon a resource decomposition scheme that separates the integer variables from the continuous ones. The decompostion of the operating problem leads to an interactive scheme which employs a mixed integer programming problem, known as the master, to generate trial operating configurations. The optimum operating conditions of each trial configuration is found using a smooth nonlinear programming model. The dual vector recovered from this model is subsequently used by the master to generate the next trial configuration. The solution algorithm progresses until lower and upper bounds converge. A range of numerical experiments are conducted and these experiments are included in the discussion. Using the operating model as a basis, a regional capacity expansion model is then developed. It determines the type, location and capacity of additional power plants and transmission lines, which are required to meet predicted electicity demands. A generalised resource decompostion scheme, similar to that used to solve the operating problem, is employed. The solution algorithm is used to solve a range of test problems and the results of these numerical experiments are reported. Finally, the expansion problem is applied to the Queensland electricity grid in Australia.

On the optimal capacity expansion of electric power systems

The analysis of investment in the electric power has been the subject of intensive research for many years. The efficient generation and distribution of electrical energy is a difficult task involving the operation of a complex network of facilities, often located over very large geographical regions. Electric power utilities have made use of an enormous range of mathematical models. Some models address time spans which last for a fraction of a second, such as those that deal with lightning strikes on transmission lines while at the other end of the scale there are models which address time horizons consisting of ten or twenty years; these usually involve long range planning issues. This thesis addresses the optimal long term capacity expansion of an interconnected power system. The aim of this study has been to derive a new, long term planning model which recognises the regional differences which exist for energy demand and which are present in the construction and operation of power plant and transmission line equipment. Perhaps the most innovative feature of the new model is the direct inclusion of regional energy demand curves in the nonlinear form. This results in a nonlinear capacity expansion model. After review of the relevant literature, the thesis first develops a model for the optimal operation of a power grid. This model directly incorporates regional demand curves. The model is a nonlinear programming problem containing both integer and continuous variables. A solution algorithm is developed which is based upon a resource decomposition scheme that separates the integer variables from the continuous ones. The decompostion of the operating problem leads to an interactive scheme which employs a mixed integer programming problem, known as the master, to generate trial operating configurations. The optimum operating conditions of each trial configuration is found using a smooth nonlinear programming model. The dual vector recovered from this model is subsequently used by the master to generate the next trial configuration. The solution algorithm progresses until lower and upper bounds converge. A range of numerical experiments are conducted and these experiments are included in the discussion. Using the operating model as a basis, a regional capacity expansion model is then developed. It determines the type, location and capacity of additional power plants and transmission lines, which are required to meet predicted electicity demands. A generalised resource decompostion scheme, similar to that used to solve the operating problem, is employed. The solution algorithm is used to solve a range of test problems and the results of these numerical experiments are reported. Finally, the expansion problem is applied to the Queensland electricity grid in Australia

A comparative study of the methods of inclusion of PMU current phasor measurements in a hybrid state estimator

The paper compares three different methods of inclusion of current phasor measurements by phasor measurement units (PMUs) in the conventional power system state estimator. For each of the three methods, comprehensive formulation of the hybrid state estimator in the presence of conventional and PMU measurements is presented. The performance of the state estimator in the presence of conventional measurements and optimally placed PMUs is evaluated in terms of convergence characteristics and estimator accuracy. Test results on the IEEE 14-bus and IEEE 300-bus systems are analyzed to determine the best possible method of inclusion of PMU current phasor measurements.

QoS-oriented sesource allocation and scheduling of multiple composite web services in a hybrid cloud using a random-key genetic algorithm

In cloud computing resource allocation and scheduling of multiple composite web services is an important challenge. This is especially so in a hybrid cloud where there may be some free resources available from private clouds but some fee-paying resources from public clouds. Meeting this challenge involves two classical computational problems. One is assigning resources to each of the tasks in the composite web service. The other is scheduling the allocated resources when each resource may be used by more than one task and may be needed at different points of time. In addition, we must consider Quality-of-Service issues, such as execution time and running costs. Existing approaches to resource allocation and scheduling in public clouds and grid computing are not applicable to this new problem. This paper presents a random-key genetic algorithm that solves new resource allocation and scheduling problem. Experimental results demonstrate the effectiveness and scalability of the algorithm.