This thesis proposes the initial stages of the development of a NUI – Asset Integrity Case (Normally Unattended Installation). An NUI – Asset Integrity Case will enable the user to determine the impact of deficiencies in asset integrity and demonstrate that integrity is being managed. A key driver for improved asset integrity monitoring is centred on the level of accurate reporting of incidents. This stems from incidents to key offshore systems and areas. For example, gas turbine driven generators where 22% of fuel gas leaks were undetected with 60% of these 22% having been found to have ignited. Accordingly, there is a need for dynamic risk assessment and improved asset integrity monitoring. The immediate objective of this research is to investigate how a dynamic risk model can be developed for an offshore system. Subsequently, two dynamic risk assessment models were developed for an offshore gas turbine driven electrical power generation system. Bayesian Networks provided the base theory and algorithms to develop the models. The first model focuses on the consequences of one component failure. While the second model focuses on the consequences of a fuel gas release with escalated fire and explosion, based upon several initiating failures. This research also provides a Multiple Attribute Decision Analysis (MADA) to determine the most suitable Wireless Sensor Network (WSN) configuration for asset integrity monitoring. The WSN is applied to the same gas turbine system as in the dynamic risk assessment models. In the future, this work can be expanded to other systems and industries by applying the developed Asset Integrity Case framework and methodology. The framework outlines the steps to develop a dynamic risk assessment model along with MADA for the most suitable remote sensing and detection methods.

The Everglades has been experiencing major changes, both climatic and anthropogenic, such that the landscape is experiencing additional stresses and forcings leading it away from its natural equilibrium. The land within and surrounding the Everglades has undergone severe modifications that may have detrimental effects on wildlife and natural features, such as rivers and landscape connectivity. Here in this study, the main focus is on understanding and quantifying hydrologic and geomorphic signatures of climatic and anthropogenic changes on the Everglades landscape. For this, in particular, available data on natural hydrological processes was used, such as rainfall, groundwater elevation, streamflow as well as surface elevations and satellite images for three different regions. These regions are categorized as forested, urban (nearby Everglades regions) and transition (in between forested and urban regions). The results show distinct differences in the statistics of observed hydrologic variables for the three different regions. For example, the probability distribution functions (PDFs) of groundwater elevation for the case of urban region show a shift in mean as well as lower asymmetry as compared to forested regions. In addition, a significant difference in the slopes between smaller and larger scales of the power spectral densities (PSDs) is observed when transitioning from forested to urban. For the case of the streamflow PDFs and PSDs, the opposite trends are observed. Basin properties extracted from digital elevation models (DEMs) of the Everglades reveal that drainage densities increase when moving from the urban to the forested sub-regions, highlighting the topographic and land use/land cover changes that the Everglades has been subjected to in recent years. Finally, computing the interarrival times of extreme (>95th percentile) events that suggest power-law behavior, the changes in power-law exponents of the hydrologic processes further highlights how these processes differ spatially and how the landscape has to respond to these changes. Quantifying these observed changes will help develop a better understanding of the Everglades and other wetlands ecosystems for management to future changes and restoration.

Apalachicola Bay in the Florida panhandle is home to a rich variety of salt water and freshwater wetlands but unfortunately is also subject to a wide range of hydrologic extreme events. Extreme hydrologic events such as hurricanes and droughts continuously threaten the area. The impact of hurricane and drought on both fresh and salt water wetlands was investigated over the time period from 2000 to 2015 in Apalachicola Bay using spatio-temporal changes in the Landsat based NDVI. Results indicate that salt water wetlands were more resilient than fresh water wetlands. Results also suggest that in response to hurricanes, the coastal wetlands took almost a year to recover while recovery following a drought period was observed after only a month. This analysis was successful and provided excellent insights into coastal wetland health. Such long term study is heavily dependent on optical sensor that is subject to data loss due to cloud coverage. Therefore, a novel method is proposed and demonstrated to recover the information contaminated by cloud. Cloud contamination is a hindrance to long-term environmental assessment using information derived from satellite imagery that retrieve data from visible and infrared spectral ranges. Normalized Difference Vegetation Index (NDVI) is a widely used index to monitor vegetation and land use change. NDVI can be retrieved from publicly available data repositories of optical sensors such as Landsat, Moderate Resolution Imaging Spectro-radiometer (MODIS) and several commercial satellites. Landsat has an ongoing high resolution NDVI record starting from 1984. Unfortunately, the time series NDVI data suffers from the cloud contamination issue. Though simple to complex computational methods for data interpolation have been applied to recover cloudy data, all the techniques are subject to many limitations. In this paper, a novel Optical Cloud Pixel Recovery (OCPR) method is proposed to repair cloudy pixels from the time-space-spectrum continuum with the aid of a machine learning tool, namely random forest (RF) trained and tested utilizing multi-parameter hydrologic data. The RF based OCPR model was compared with a simple linear regression (LR) based OCPR model to understand the potential of the model. A case study in Apalachicola Bay is presented to evaluate the performance of OCPR to repair cloudy NDVI reflectance for two specific dates. The RF based OCPR method achieves a root mean squared error of 0.0475 sr?1 between predicted and observed NDVI reflectance values. The LR based OCPR method achieves a root mean squared error of 0.1257 sr?1. Findings suggested that the RF based OCPR method is effective to repair cloudy values and provide continuous and quantitatively reliable imagery for further analysis in environmental applications.

A mathematical model is derived to provide quasi-steady state predictions of the performance of a high pressure hydraulic supply system, using equations which govern the physical processes as opposed to equations which match input-output characteristics. Model equations are developed to describe the operation of the power source, control valves, energy source, gas side of the system, hydraulic accumulator, the motorpump, and hydraulic side of the system. The accuracy of the model is then checked by inserting known parameters from a previously developed control system and comparing model predictions with performance data from this system.

Heat energy is important to all aspects of life. Various industries including food processing plants, chemical processing plants, thermal power plants, refrigeration and air conditioning equipments, petrochemical plants, etc. are faced with the problems of effective utilization, conservation and recovery of heat. The production of heat exchangers involves huge investments for capital and operation costs. In view of this, it has become important to design heat exchangers that will be efficient and also save energy, cost and materials. Different techniques known as heat transfer enhancement are employed to achieve this. Of these techniques, the tube-insert technology is applied in the present research. No prior work on heat transfer and fluid flow in a tube induced with twisted tape insert with emphasis on cuts with different geometrical shapes but equal area has been reported. Hence, in the present work, heat transfer and fluid flow of water in tubes induced with twisted tape inserts with different-shape-equal-area cuts is investigated numerically. The present studies pay attention to the thermo-hydraulic characteristics of laminar, transitional and turbulent flows of water through different tube designs fitted individually with twisted tape of different design under uniform wall heat flux. The numerical simulation in this work is carried out by using Fluent software. The RANS-based RNG κ-ε model is employed for the turbulent flow because it is found to give a more accurate result than other turbulence models. Since transitional flow is not fully turbulent, the transitional variant of the SST κ-ω model is applied for the simulation of the transitional flow. The analyses quantify the improvement in the heat transfer, friction factor and thermal performance index in each of the tube systems and these results are used to ascertain the system that gives the best performance. Correlations are also proposed for the Nusselt number and friction factor. The results indicates that the superior fluid mixing provided by the alternate-axis triangular cut twisted tape is one of the reasons why it offers heat transfer enhancement and thermal performance factor that is higher than those that are offered by other induced tubes. Interestingly, the enhancement in heat transfer increases as the size of the cuts on the tape and the width of the tape increase but decreases as the pitch of the tape increases. The enhancement of heat transfer affects the start and the end of the transition to turbulent flow. Transition to turbulent flow occurs and ends earliest in the tube system with the highest heat transfer enhancement. Investigation is also performed on the combined forced and free convection heat transfer in an inclined tube for laminar, transitional and turbulent flows. The induced tube is inclined at different angles (15o≤θ≤90o ) with respect to the horizontal. Importantly, the heat transfer enhancement of the tube systems under mixed convection is higher than those under forced convection, and the enhancement for the mixed convection increases as the inclination angle increases.

Nanofiltration (NF) membranes are used for a range of industrial applications one of which is for the removal of the sulphate constituent in seawater. This is a mature activity for the treatment of seawater that is to be injected into oil reservoirs in the offshore oil/gas industry. Such sulphate removals have also been the subject of much interest, as a pretreatment strategy, in seawater desalination plants that is either utilising thermal technology or reverse osmosis. Nevertheless, there is a need for robust criteria, such as the comparative permeate flux and sulphate rejection, of selecting the optimum NF membrane. There is a major difficulty in the assessment of the comparative filtration performance and the role of membrane structure because the data from manufacturers and also the information from the scientific literature emanates from different testing protocols. This can result in an enigmatical situation for obtaining the optimum NF membrane for a particular application. Against the above background this PhD project has focused on undertaking a fundamental study of different commercially available NF membranes in order to facilitate improved assessment of their filtration performance for sulphate rejection applications in relevant standardised testing conditions. Moreover, on the basis that those variations in membranes’ functioning are attributed to membrane structure and characteristics, a major segment of the research was focused on correlating filtration performance and membrane features. The research comprised two main phases; the first phase involved determining the comparative filtration performance of eight commercially available NF membranes supplied from four manufacturers. The second main phase was to undertake detailed characterisation studies on the NF membranes in order to obtain a clear understanding of their sulphate separation mechanism and permeate flux. The first phase involved assessments of the permeate flux and selectivity of the eight membranes. The experimental protocol in the second, characterisation part of the work was directed to the identification and evaluation of NF active surface layer characteristics:  Pore characterisation by porosity factor calculations,  Hydrophilicity/Hydrophobicity nature by contact angle measurements,  Surface Free Energy calculations,  Surface roughness measurements by AFM,  Membrane potential measurements and average charge density calculations. This approach is an acknowledged strategy for NF membrane scientific research assessment and, in the current work provided key data of membrane features that facilitated a systematic understanding of membrane functioning. These characterisation features were also linked successfully to the membrane performance parameters to yield a characterisation/performance envelope which represents a useful basis for NF membrane selection and utilisation to optimise membrane usage and consequent economic advantage. The general discussion includes a summary of the interface between the role of NF and the operational and economic features of the two main types of desalination processes. It includes an outline of a process scheme for the incorporation of NF pretreatment into an MSF plant from the conceptual design stage as opposed to the application employed hitherto where the emphasis has been on attaching NF pretreatment equipment on to an existing unit. As a result, it is expected that NF usage should increase performance ratio, reduce energy consumption, hence the running cost, and increase recovery.

Pipeline distribution systems account for the vast majority of the physical infrastructure in the water industry. Their effective management represents the primary challenge to the industry, from both an operational and public health standpoint. Biofouling has a ubiquitous presence within these systems, and it can significantly impede their efficiency, through an increase in boundary shear caused by characteristic changes in surface roughness dynamics. Nonetheless, conventional pipeline design practices fail to take into account such effects, partially because research findings that could contribute to upgraded and optimised design practices appear inconsistent in the literature. The overall aim of this study was to improve the current scientific understanding of biofouling within water and wastewater pipelines; for the purpose of instigating a step-change in pipeline design theory by incorporating biofouling, thereby enabling future pipelines to be as sustainable as possible. The nature of the problem, necessitated the need for a multidisciplinary approach, based upon engineering and microbiological principles and techniques. The primary focus of this study was to investigate the impact of biofouling on surface roughness, mean flow structure and sediment transport within wastewater systems. To this effect biofilms were incubated with a synthetic wastewater on a High Density Polyethylene pipe, within a purpose built pipeline facility for 20 days, at three steady-state flow regimes, including the average freestream velocities of 0.60, 0.75 and 1.00 m/s. The physico-chemical properties of the synthetic wastewater were purposely designed to be equivalent to the properties associated with actual wastewater found within typical European sewers. The impact of biofouling on flow hydrodynamics was comprehensively identified using a series of static pressure tappings and a traversable Pitot probe. Molecular and image analysis was also undertaken to support the observations derived from the aforementioned measurements, particularly with regards to the structural composition and mechanical stability of the biofouled surfaces. The study has confirmed that the presence of a low-form gelatinous biofilm can cause a significant increase in frictional resistance and equivalent roughness, with increases in friction factor of up to 85% measured over the non-fouled values. The reported increases in frictional resistance resulted in a reduction in flow rate of up to 22% and increased the pipe’s self-cleansing requirements. The structural distribution of a biofilm was shown to play a key role in its overall frictional capacity and strength, which in turn was found to be a function of the biofilms conditioning shear. In particular, it was found that a biofilm conditioned at higher shear will have less of an impact on a pipe’s overall frictional resistance, although, will be stronger and more difficult to remove than a biofilm conditioned at lower shear. The biofilm’s impact on frictional resistance was found to be further compounded by the fact that traditional frictional relationships and their derivatives are not applicable to biofouled surfaces in their current manifestation. In particular, the von Kármán constant, which is an integral aspect of the Colebrook-White equation is non-universal and dependent on Reynolds Number for biofouled surfaces. It was found that the most suitable manner to deal with the dynamic and case-specific nature of a biofouled surface was to quantify it using a series of dynamic roughness expressions, the formulation of which were the culmination of this study, and should be the focus for further research. The influence of different plastic based pipe materials and flow regimes on biofilm development within drinking water distribution systems was also briefly investigated using a series of flow cell bioreactors and molecular analysis techniques. Keywords: Biofilm; biofouling; pipe; hydraulic efficiency; equivalent roughness; von Kármán constant; Colebrook-White equation; drainage network; wastewater; drinking water.

A new approach to solve shallow water flow problems over highly irregular geometry both correctly' and efficiently is presented in this thesis. Godunov-type schemes which are widely used with the finite volume technique cannot solve the shallow water equations correctly unless the source terms related to the bed slope and channel width variation are discretized properly, because Godunov-type schemes were developed on the basis of homogeneous governing equations which is not compatible with an inhomogeneous system. The main concept of the new approach is to avoid a fractional step method and transform the shallow water equations into homogeneous form equations. New definitions for the source terms which can be incorporated into the homogeneous form equations are also proposed in this thesis. The modification to the homogeneous form equations combines the source terms with the flux term and solves them by the same solution structure of the numerical scheme. As a result the source terms are automatically discretized to achieve perfect balance with the flux terms without any special treatment and the method does not introduce numerical errors. Another point considered to achieve well-balanced numerical schemes is that the channel geometry should be reconstructed in order to be compatible with the numerical flux term which is computed with piecewise constant initial data. In this thesis, the channel geometry has been changed to have constant state inside each cell and, consequently, each cell interface is considered as a discontinuity. The definition of the new flux related to the source terms has been obtained on the basis of the modified channel geometry. A simple and accurate algorithm to solve the moving boundary problem in two-dimensional modelling case has also been presented in this thesis. To solve the moving boundary condition, the locations of all the cell interfaces between the wet and dry cells have been detected first and the integrated numerical fluxes through the interfaces have been controlled according to the water surface level of the wet cells. The proposed techniques were applied to several well-known conservative schemes including Riemann solver based and verified against benchmark tests and natural river flow problems in the one and two dimensions. The numerical results shows good agreement with the analytical solutions, if available, and recorded data from other literature. The proposed approach features several advantages: 1) it can solve steady problems as well as highly unsteady ones over irregular channel geometry, 2) the numerical discretization of the source terms is always performed as the same way that the flux term is treated, 3) as a result, it shows strong applicability to various conservative numerical schemes, 4) it can solve the moving (wetting/drying) boundary problem correctly. The author believes that this new method can be a good option to simulate natural river flows over highly irregular geometries.

This thesis takes top tensioned vertical riser interaction as its main study object. It has its focus on the understanding of the mechanism about potential instability of the risers caused by the interaction and the prediction of potential collision. Started from two-dimensional cylinder interaction cases, and later extended into the three-dimensional riser scenarios, the problem is investigated comprehensively. The study covers fluid force prediction, stability analysis, continuation investigation and dynamics simulation. The study disclosed the mechanism of the potential collision when the flow velocity reaches a certain critical value, and provides a robust tool to predict the threshold for the riser collision. Additionally, the investigation shows the difference between marine riser interaction and the similar interaction occurs in other engineering disciplines, such as power transmission lines, heat exchangers etc. Also provided in this thesis are valuable information regarding the riser dynamics should collision occur. The research will be beneficial to the marine riser designers and operators.

The design aspects of offshore jacket structures are presented and discussed with a special emphasis on the different factors which affect wave loading calculations for these structures. An up-to-date review of a large amount of data on the hydrodynamic coefficients obtained from Laboratory experiments and wave projects is presented and the main data are tabulated. To assess the different aspects of the wave loading a set of computer programs were developed and used to perform various comparative studies for the existing methods of wave loading estimation. The analysis of the wave loading was carried out using a jacket structure of 119 members having 73m x 73m base representing a typical offshore platform, assumed to be working in 150m of water. The general method of wave loading calculation is based on Morison's equation taking into account the phase differences between the velocities and accelerations of the wave particles. The relative positions of the different members in space and time when the wave passes through the jacket were also considereG. Besides the drag and inertia forces, the lift (transverse) forces are also taken into account. The kinematics of the flow can be determined using Airy (linear) wave theory, Stokes 2nd order theory or Stokes 5th order theory. Constant drag and inertia coefficients (CD, CM), as recommended by Lloyd's Register of Shipping (LR), Det Norkse Veritas (DnV) and Bureau Veritas (I3V),can be used. Alternatively, variable hydrodynamic coefficients (CD, CM, CL) from Sarpkaya's experimental data for smooth and rough cylinders can be used. The drag interference effect and the current effect can be included in the calculations. Various interpretations as to how to apply Morison's equation in the design were examined which have shown the importance of taking full account of both the relative positions in space and time of the different members of the structure as well as the phase relationships in the wave. A comparison was made between the results of calculations using the recommended coefficients (CD, CM) of LR, DnV and BV which has shown that even small variations in these coefficients leads to appreciable differences in the loading estimation of up to 45%. The approach using variable coefficients (Sarpkaya's data), which are related to the local Reynolds number (Re) and Keulegan-Carpenter number (K) at the different points of the structure, was compared with the method of adopting constant coefficients (as recommended by LR) showed differences up to 26% in the wave loading estimation between the two methods. The effects of surface roughness, as well as the transverse (lift) forces, on the wave loading were also investigated and found to be very significant (eg 43% to 56% in the surge force) and should be considered in design. Three wave theories (Airy, Stokes 2nd order, Stokes 5th order) were compared in terms of wave profile, horizontal and vertical velocities and accelerations. The results have shown that the differences in predicting the wave kinematics by Airy and Stokes theories are large. The wave forces on the individual members as well as the total forces and moments on the complete structure calculated by the fifth order theory, showed 30-60% differences when compared with the results based on Airy theory. The experimental data on the interference effect between the cylindrical members were reviewed. The effect on the jacket loading was examined using some experimental data and found to be 6-9% reduction in the loading for rough cylinders. However, more experimental investigations are required in this area to deal with this problem properly. The effect of current speed and direction on the wave loading was examined by the commonly used practice of adding the velocity of current vectorially to the wave particle velocity when calculating the drag and lift forces. The results showed that the total forces and moments could be increased by 16-37% for a/1 mls current in the direction of the wave. Several static analyses of the jacket were performed using constant and variable hydrodynamic coefficients and two wave theories (Airy and Stokes 5th order theory). The initial differences in the wave loading due to the different coefficients and wave theories appeared again as appreciable differences in the maximum stress on the different members. This supported the necessity of calculating the wave loading accurately from the beginning. A general review of the reliability analysis method as applied to jacket structures indicated that the modelling of the wave loading needs further improvements to take account of the large uncertainties in the loading especially due to the hydrodynamic coefficients and non-linear loads.