Photoplethysmography (PPG) is a non-invasive photometric technique which measures changes in the volume of blood in the biological tissue. PPG is well-known for its application in pulse oximetry used for the continuous monitoring of arterial blood oxygen saturation (SpO2). Over the past decade, there has been a plethora of research in the field of PPG, with potential applications beyond pulse oximetry and heart rate monitoring. Such applications explore the utilisation of PPG for the assessment of various bio-markers relating to vascular mechanics, haemodynamics and many others. With the growing research interest in the field of PPG, a comprehensive understanding of the light-tissue interaction-based working principle underlying the technique is essential. This thesis is focussed on the investigation of the fundamental light-tissue interactions in PPG using the Monte Carlo method. Tissue models have been developed in this thesis which were characterised by the optical properties (e.g., wavelength- dependent coefficients of scattering and absorption etc.), the anatomical features (e.g., stratification and dimension of tissue layers and sublayers etc.), and the physiological parameters (water and blood content in tissue layers etc.). The Monte Carlo strategy was verified, and was initially implemented to model the light propagation through a monolayer perfused dermal tissue volume in a reflective mode PPG at the red and near-infrared wavelengths, usually used in pulse oximetry. Results illustrated the distribution of the scattering-absorption interaction events, and quantified the optical pathlength, penetration depth and detected reflectance with the variable sensor geometry (i.e., source-detector separation) and physiological states (i.e., the volume of blood and oxygen saturation) of the tissue. The monolayer model was also employed to produce the plot resembling the 'calibration curve' used in pulse oximetry. With the knowledge gained from the monolayer-model study, a similar investigation was performed on a heterogeneous tissue structure of a human finger which was executed in both reflective and transmissive geometrical settings. The calibration curves produced from the detected reflectance and transmittance exhibited a high correlation. The absorbances of red and near-infrared light by individual layers of the finger were quantified at systole and diastole. To the relative absorbance, the contributions of dermis and bone were the maximum and the minimum, respectively. The dependence of the optical pathlength on the source-detector separation and the operating wavelength was quantified by the Differential Pathlength Factor (DPF), which was assessed for the reflective mode PPG by simulating light propagation through a human forearm tissue volume. The DPF values were used in experimentally obtained PPG signal in order to determine the time-change in the concentration of oxyhaemoglobin and deoxyhaemoglobin. Cross-talk and absolute errors were calculated between the simulated and approximated DPFs. The results presented in the thesis contribute greatly to the understanding on PPG light-tissue interaction. Such knowledge could also greatly contribute to the development of the new generation PPG sensors for various applications.

This research aims to study in a comprehensive way a different power generation cryogenic energy storage cycles and effective strategies for developing an optimized design of small scale nitrogen axial turbines as the expanders for these cycles within the capacities that can be used for small/medium size buildings, rural, and remote off-grid communities. The hybrid open-closed Rankine cycle have been chosen as the case study for nitrogen turbine analysis for expansion ratios ranged from 1.5 to 3. New turbine design methodology has been developed which integrates one dimension preliminary design method (mean-line method) and three dimensional CFD simulations, and expe1imental validation testing. This turbine methodology was expanded to include developing optimization parametrization technique, a parametric study of four different blade configurations (lean, sweep, twist, and bow), and development of a novel dual stage non-repeated annular area small-scale axial nitrogen turbine. In order to validate the CFD simulation, the design methodology, and to investigate the effects of blade height on small-scale axial turbines performance, a test rig using compressed air was developed. Three manufactured axial turbines with different blade heights ( 4mm, 6mm, and 8mm) were manufactured and tested at various operating conditions.

Focusing on the author's research aspects, the intelligent optimization algorithm and advanced control methods of the diesel engine's air path have been proposed in this work. In addition, the simulation platform and the HIL test platform are established for research activities on engine optimization and control. In this thesis, it presents an intelligent transient calibration method using the chaos-enhanced accelerated particle swarm optimization (CAPSO) algorithm. It is a model-based optimization approach. The test results show that the proposed method could locate the global optimal results of the controller parameters within good speed under various working conditions. The engine dynamic response is improved and a measurable drop of engine fuel consumption is acquired. The model predictive control (MPC) is selected for the controllers of DLEGR and VGT in the air-path of a diesel engine. Two MPC-based controllers are developed in this work, they are categorized into linear MPC and nonlinear MPC. Compared with conventional PIO controller, the MPC-based controllers show better reference trajectory tracking performance. Besides, an improvement of the engine fuel economy is obtained. The HIL test indicates the two controllers could be implemented on the real engine.

This thesis investigates the feasibility of producing electricity and cooling simultaneously utilising low-grade heat sources by incorporating an expander within the adsorption cooling system or by integrating an Organic Rankine Cycle with water adsorption cooling system. Advanced physical adsorbent materials have been investigated for the first time to generate cooling and electricity simultaneously utilising CPO-27(Ni), MIL101(Cr), and AQSOA-Z02 and compared to commonly used Silica-gel. Two innovative configurations of water adsorption systems for cooling and electricity were investigated. In the first configuration, the two-bed basic adsorption cooling system (BACS) is improved by including an expander within the system. In the second configuration, the BACS and ORC cycle are integrated. Four different scenarios of systems integration based on the way of powering the ORC and the adsorption system were investigated. Also, detailed CFD simulations of small-scale radial inflow turbines are developed for both configurations. Also, a novel experimental facility is developed to integrate ORC with two-bed adsorption cooling system to validate the numerical models and proof the concept of producing power as well as cooling, where maximum specific cooling power of 252 W/kgads and specific power and of 162 W/kgads can be achieved with maximum deviation of less than 17%.

This research introduces two different methods that are Levy Flight and Hooke and Jeeves to the Bees Algorithm with the aim of improving the convergence speed and its robustness. Both methods are incorporated to the Bees Algorithm at neighbourhood search of the elite bees since that particular locations are the most promising area during optimisation process. Each Bees Algorithm and the newly incorporated method with thirteen different parameter settings are subjected to fifteen different benchmark test functions. These benchmark test functions are represented with different characteristics in terms of its differentiability, separability, scaleability, and modality. Bees Algorithm with Levy-flight method incorporated to the local search performs excellent result for 13 out of 15 functions against standard Bees Algorithm in terms of its success rate and convergence speed in which it is validated by the statistical T test. As a matter of fact, the new method indicates better robustness for 13 functions in terms of achieving good result for solving different types of optimisation problems. For the Bees Algorithm with Hooke and Jeeves method, the new approach reaches a relatively better performance compared with standard Bees Algorithm in which one parameter excels at reaching optimum solution for most of the test functions.

Digitized by Kansas Correctional Industries

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Digitized by Kansas Correctional Industries

Digitized by Kansas Correctional Industries

Digitized by Kansas Correctional Industries