Mathematical models are often used to describe and, in some cases predict, excitable cellular behaviour that is based on observed experimental results. With the increase of computational power, it is now possible to solve such models in a relatively short time. This, along with an increasing knowledge of cellular and subcellular processes, has led to the development of a large number of complex cellular models, capable of describing a broad range of excitable cell behaviour. But the use of complex models can also lead to problems. Most models can accurately reproduce results associated with the data on which the models are based. However, results from complicated models, with large numbers of variables and parameters, are less reliable if the model is not placed under the same physiological conditions as defined by the model author. In order to test a model??s suitability and robustness over a range of physiological conditions, one needs to fit model parameters against experimental data observed under those conditions. By using the modelling standard and repository offered by CellML, model users can easily select and adapt a large number of models to set up their own applications to fit model parameters against user-supplied experimental data. However, currently there is a lack of software that can utilise CellML model for parameter fitting. In this thesis, a Java-based utility has been developed, capable of performing least square parameter optimisation for a wide range of CellML models. Using the developed software, a number of parameter fits and identifiability analyses were performed on a selected group of CellML models. It was found that most of the models were ill-formed, with larger numbers of parameters worsening model identifiability. In some cases, the usage of multiple datasets and different objective functions can improve model identifiability. Finally, the developed software was used to perform parameter optimisation against two sets of action potentials from a sinoatrial node experiment, in the absence and presence of E9031, a specific ion channel blocker.

Mathematical models are often used to describe and, in some cases predict, excitable cellular behaviour that is based on observed experimental results. With the increase of computational power, it is now possible to solve such models in a relatively short time. This, along with an increasing knowledge of cellular and subcellular processes, has led to the development of a large number of complex cellular models, capable of describing a broad range of excitable cell behaviour. But the use of complex models can also lead to problems. Most models can accurately reproduce results associated with the data on which the models are based. However, results from complicated models, with large numbers of variables and parameters, are less reliable if the model is not placed under the same physiological conditions as defined by the model author. In order to test a model??s suitability and robustness over a range of physiological conditions, one needs to fit model parameters against experimental data observed under those conditions. By using the modelling standard and repository offered by CellML, model users can easily select and adapt a large number of models to set up their own applications to fit model parameters against user-supplied experimental data. However, currently there is a lack of software that can utilise CellML model for parameter fitting. In this thesis, a Java-based utility has been developed, capable of performing least square parameter optimisation for a wide range of CellML models. Using the developed software, a number of parameter fits and identifiability analyses were performed on a selected group of CellML models. It was found that most of the models were ill-formed, with larger numbers of parameters worsening model identifiability. In some cases, the usage of multiple datasets and different objective functions can improve model identifiability. Finally, the developed software was used to perform parameter optimisation against two sets of action potentials from a sinoatrial node experiment, in the absence and presence of E9031, a specific ion channel blocker.

<p>The goal of this thesis has been to study the behaviour of the closed loop driver-vehicle-environment in simulation and to find parameters of the synthetic vehicle model, which minimise certain optimisation criteria. A method of optimising parameters using genetic algorithms has been implemented and has proven to work well. Two different driving strategies have been tried in the optimisation of an ISO lane-change maneouvre. The first approach has simulated a beginner driver and his or her behaviour. The second approach simulates an experienced driver and also the possibility of driver adaption to different vehicle types. The implemented driver model has shown to be sufficient to describe the driver's behaviour during lateral maneouvres. A parameter set which minimises the lateral acceleration response on steering wheel angle has proven to be the optimum. This includes a small steering wheel ratio, and a small but positive under steer gradient. The driver has demonstrated the ability to adapt to different vehicles, and therefore different parameter sets, describing the driver, should be used for different problems.</p>

The goal of this thesis has been to study the behaviour of the closed loop driver-vehicle-environment in simulation and to find parameters of the synthetic vehicle model, which minimise certain optimisation criteria. A method of optimising parameters using genetic algorithms has been implemented and has proven to work well. Two different driving strategies have been tried in the optimisation of an ISO lane-change maneouvre. The first approach has simulated a beginner driver and his or her behaviour. The second approach simulates an experienced driver and also the possibility of driver adaption to different vehicle types. The implemented driver model has shown to be sufficient to describe the driver's behaviour during lateral maneouvres. A parameter set which minimises the lateral acceleration response on steering wheel angle has proven to be the optimum. This includes a small steering wheel ratio, and a small but positive under steer gradient. The driver has demonstrated the ability to adapt to different vehicles, and therefore different parameter sets, describing the driver, should be used for different problems.

Material utilisation is of importance in the manufacturing industry formaking the most of each material, minimising waste and increasing cost-effectiveness. In this thesis, samples of Waspaloy built with the method of L-DED-P has been analysed in order to investigate how process pa-rameters influence the build quality and geometrical accuracy in as-builtobjects. The samples analysed was built in single rows of one, three,five and fifteen layers. A build process was used in which the sampleswere built with individual combinations of the process parameters laserpower, scanning speed, and powder flow. Each combination of processparameters was used to build one track for each layer height.Analysis included defect analysis with light optical microscopy, andpost-processing with ImageJ for automatic identification, quantification,and collection of measurements. A qualitative analysis was performed andthe sample properties and characteristics was described in terms of theamount of defects, including a descriptive assessment of defect severity.Etched samples revealed a columnar grain structure in samples, which was apparent in builds with at least three layers.The results presented show a difference in build height, quantity andsize of pores, and the presence or absence of other defects such as lack of fusion. Sample 3 which was built with high laser power, slow scan-ning speed, and high powder feed show promising results with one ofthe highest build rates of all samples, combined with a low normalisedpore distribution. The sample experiences partial hardening, with hard-ness values reaching 320 HV, but still promisingly show no sign of crackformation.It is concluded that powder feed relates primarily to the build rateof the samples, and the scanning speed together with the laser power influence the quality of the build, where high laser power and low scanningspeed tends to form well behaving samples with few defects, whilst other combinations increase the risk of defects.

In mammalian hearts, initiation of the heartbeat occurs in a region of specialised pacemaker cells known as the sinoatrial node (SAN). The SAN is a highly complex spatially distributed structure which displays considerable cellular heterogeneity and is subject to complex electrotonic interactions with the surrounding atrial tissue. In this study, biophysically detailed ionic models of central and peripheral SAN pacemaker cells are described. These models are able to accurately reproduce experimental recordings of the membrane potential from central and peripheral SAN tissue. These models are used to investigate frequency synchronisation of electrically coupled cardiac pacemaker cells. Based on simulation results presented, it is proposed that cellular heterogeneity in the SAN plays an important role in achieving rhythm coordination and possibly contributes to the efficient activation of the surrounding atrial myocardium. This represents an important, previously unexplored, mechanism underlying pacemaker synchronisation and cardiac activation in vivo. A spatial-gradient model of action potential heterogeneity within the SAN is then formulated using a large-scale least squares optimisation technique. This model accurately reproduces the smooth spatial variation in action potential characteristics observed in the SAN. One and two dimensional models of the intact SAN are then formulated and three proposed models of SAN heterogeneity are investigated: 1) the discrete-region model, in which the SAN consists of a compact central region surrounded by a region of transitional pacemaker cells, 2) the gradient model, in which cells of the SAN exhibit a smooth variation in properties from the centre to the periphery of the SAN, and 3) the mosaic model, in which SAN and atrial cells are scattered throughout the SAN region with the proportion of atrial cells increasing towards the periphery. Simulation results suggest that the gradient model achieves frequency entrainment more easily than the other models of SAN heterogeneity. The gradient model also reproduces action potential waveshapes and a site of earliest activation consistent with experimental observations in the intact SAN. It is therefore proposed that the gradient model of SAN heterogeneity represents the most plausible model of SAN organisation.

The quadratic class of time-frequency distributions (TFDs) forms a set of tools which allow to effectively extract important information from a nonstationary signal. To determine which TFD best represents the given signal, it is a common practice to visually compare different TFDs' time-frequency plots, and select as best the TFD with the most appealing plot. This visual comparison is not only subjective, but also difficult and unreliable especially when signal components are closely-spaced in the time-frequency plane. To objectively compare TFDs, a quantitative performance measure should be used. Several measures of concentration/complexity have been proposed in the literature. However, those measures by being derived with certain theoretical assumptions about TFDs are generally not suitable for the TFD selection problem encountered in practical applications. The non-existence of practically-valuable measures for TFDs' resolution comparison, and hence the non-existence of methodologies for the signal optimal TFD selection, has significantly limited the use of time-frequency tools in practice. In this thesis, by extending and complementing the concept of spectral resolution to the case of nonstationary signals, and by redefining the set of TFDs' properties desirable for practical applications, we define an objective measure to quantify the quality of TFDs. This local measure of TFDs' resolution performance combines all important signal time-varying parameters, along with TFDs' characteristics that influence their resolution. Methodologies for automatically selecting a TFD which best suits a given signal, including real-life signals, are also developed. The optimisation of the resolution performances of TFDs, by modifying their kernel filter parameters to enhance the TFDs' resolution capabilities, is an important prerequisite in satisfying any additional application-specific requirements by the TFDs. The resolution performance measure and the accompanying TFDs' comparison criteria allow to improve procedures for designing high-resolution quadratic TFDs for practical time-frequency analysis. The separable kernel TFDs, designed in this way, are shown to best resolve closely-spaced components for various classes of synthetic and real-life signals that we have analysed.