Thesis (Sc. D.)--Massachusetts Institute of Technology, Dept. of Civil and Sanitary Engineering, 1935. / Vita. / Includes bibliographical references (leaf 102). / by Rolf Eliassen. / Sc.D.

With the rapid growth of the penetration of wind power into the power system, fixed series compensation is considered as an economic solution to increase power transfer capability. This will render the power system vulnerable to Sub-Synchronous Resonance (SSR). This thesis conducts research on the effect of wind turbines represented by Fixed Speed Induction Generator-Based Wind Turbines (FSIG-WTs) and Fully Rated Converter-Based Wind Turbines (FRC-WTs) on damping SSR. Firstly, SSR is investigated through mathematically modelling IEEE First Benchmark Model (FBM) using MATLAB package. Modal analysis is used to study SSR over a wide range of series compensation percentages. Secondly, the effect of incorporating FSIG-WTs into FBM on SSR is studied over a wide range of series compensation percentage and different power size of FSIG-WTs. Furthermore, the ability of the grid-side converters of the FRC-WTs connected with the FBM to damp SSR occurrence in the steam turbine shafts is evaluated using two different types of control. An optimal controller based on a Linear Quadratic regulator (LQR) has been designed as an auxiliary controller of the grid-side converter of FRC-WTs. A full-order observer was designed to estimate the unmeasured state variables to enable a vii full-state feedback. Finally, eigenvalue sensitivity was studied to choose the most suitable feedback signal for an SSR damping controller. Lead/Lag compensation controller based on the residue method is designed as an auxiliary controller within the grid-side converters of FRC-WTs. Eigenvalue analysis and time domain simulations over widely varying levels of series compensation have been carried out. The simulation studies were carried out in MATLAB and PSCAD. Connecting FSIG-WTs to the FBM increases the range of series compensation level at which SSR can occur. Therefore, it was shown that FSIG-WTs have an adverse effect on the SSR occurring at the multi-mass synchronous generator. If the system is visible, LQR as an auxiliary damping controller within the grid-side converters of FRC-WTs is an effective controller to damp SSR over a wide range of series compensation percentages. Based on eigenvalue sensitivity technique, synchronous generator speed deviation is the most suitable feedback signal for damping SSR occurrence in the steam turbine shafts.

In this EngD project I developed a new, analytic equation of state for use in the area of CCS pipeline transport. It was my aim to design a model which would exhibit a high degree of accuracy within the anticipated window of operation of CCS pipelines; from 260 to 335K and 1 to 200bar, whilst simultaneously retaining a simplicity and ease-of-use, a lack of which made some other available equations particularly unwieldy. Having conducted a comprehensive literature review and attended many academic and industrial conferences throughout this project, I felt that there was a need for an equation of state which could perform both these functions. This was the key motivation for my work, and the model presented in this thesis was developed in order that it might contribute towards negating the many concerns that currently surround the pipeline transport stage of CCS. I aimed for the proposed model to display a complexity approaching that of some of the simpler equations currently available, whilst incorporating sucient flexibility to give thermodynamic predictions to a standard approaching that of those which are more complicated. I defined criteria by which the proposed model could be judged, so that it could be applied with condence in the determination of the physical properties of carbon dioxide mixtures during CCS pipeline transport. Work was carried out by fitting the parameters of the proposed model to experimental data gathered from the literature, so that it would be able to determine the homogeneous phase pressure and vapour-liquid equilibrium behaviour of carbon dioxide and other relevant gas mixtures. The project yielded a number of excellent outputs, not least the satisfaction of the primary aim which was the proposal of a model, which through this EngD, I demonstrated had the ability to meet the demands that were set. In carrying out this work, I also developed several highly useful auxiliary mathematical methods which helped in ensuring the proposed model was as accurate as possible. For the case of modelling pure carbon dioxide, the proposed equation worked exceptionally well, providing highly accurate predictions for homogeneous density and vapour liquid equilibrium, which were well within the targets set. A paper on this was published in May 2013. In extending the model to incorporate some binary mixtures I again found that it demonstrated a clear ability to capture the necessary physical behaviours within the target range. I concluded with suggestions as to ways in which the work presented here could be developed further, as well as the many avenues for future work in other areas that this EngD project had opened up.

The main aim of this research is to investigate occupants’ thermal comfort and energy performance of multi-storey residential buildings in one of the major cities in Libya (Darnah). The work was implemented in stages as follows: 1. Gathering and analysing real weather data from a number of locations in Libya. 2. Gathering and analysing building data from residential multi-storey blocks in Darnah. 3. Conducting preliminary computer analyses using the above information to get a better idea about thermal conditions inside multi-storey block flats in Darnah. 4. Visiting Darnah and collecting real data using specifically designed survey or questionnaire forms to understand and assess how people feel while living inside their flats. 5. Analysing the computer results and field data and trying to come up with solutions to improve existing multi-storey blocks in terms of indoor climate and energy consumption and introducing guidelines for designers of new buildings in Libya. The subjective data was collected and tabulated by using a questionnaire, which has been widely used and shown to be effective, to determine people's responds through questions modified especially for the comfort purpose. Questionnaires were collected from households of 12 buildings: involving a total of 337 flats. The questionnaires compare the significance of the thermal sensation, the thermal comfort, and the building design. The results show that the thermal comfort is not satisfied, unless modified, and improving take place in these case study buildings. Thus a modification is proposed, the results from the present study show also that the insulation material is essential in this type of climate and can help to reduce up to 63% of heat gain and a reduction of 6 degrees of indoor temperature. Results also indicate that the construction of residential dwellings using existing materials is less conducive to the climatic conditions of hot dry climates and not suitable for the occupants’ requirements of comfort. Human thermal comfort was assessed using the adaptive model, to show that the climate and building’s envelope have a significant impact on human comfort perception and indoor environment. This research is mainly aimed at "explaining" the trends in the energy flows within buildings under the climatic conditions of Mediterranean zones and to establish a comfort zone for occupants within this type of climate. The results should help in the formulation of design guidelines for use in the process of building design by builders, architects and engineers. Also this research is in line with the Libyan’s government aim to make better use of renewable energy sources such as the sun for keeping buildings comfortable for their occupants.

The continuing need for improved assessment of risk from heavy metal contamination of the environment has prompted scientific interest in quantifying and predicting metal solubility, ‘lability’ and bioavailability. This has led to the development of new techniques to fractionate and speciate trace metals in soils. The objectives of the current study were to increase understanding of the effects of (i) soil properties, (ii) contaminant source and (iii) contact time on metal lability and solubility in soils. Multi-stable isotope dilution (ID) methods were used to determine the lability (E-values) of Ni, Cu, Zn, Cd and Pb in soils, alongside more traditional approaches employed for metal fractionation including single and sequential extraction procedures. Most of the work was undertaken using (i) archived soils amended by metal salts (MA soils; n=23) and (ii) topsoils collected from Nottingham, Wolverhampton and London (Urban soils; n=100). The resulting data was used first to quantify the factors affecting trace metal lability in the MA soils using a logistic (S-shape) model which described metal E-value primarily as a function of soil pH with secondary influences from other soil properties. It was apparent that mineral oxides were important fixation phases for Ni, Zn and Cd while Pb was strongly affected by organic matter in soils. This model, parameterised on soils contaminated originally by metal nitrate solution, was then applied to the Urban soils to reveal the extent to which contaminant source still controlled metal lability. A further investigation of the long-term effect of metal source on metal lability was pursued through a third dataset of rural roadside soils (n=42) which had received Pb mainly from petrol-derived and geogenic sources, defined by their isotopic signature. It was demonstrated that petrol-derived Pb remained more labile than Pb from the parent material, despite decades of contact, although both petrol-derived and geogenic Pb contributed to both the labile and non-labile fractions. In a fourth dataset, soils that had received Pb from sewage sludge amendment (n=16), the co-existence of high phosphate concentration from sewage sludge limited the magnitude and range of Pb lability, probably through formation of Pb-phosphate minerals. No consistent agreement was found between labile fraction of Pb and any single sequential extraction (SEP) fraction in all soils contaminated by Pb from multiple sources. Both empirical equations (extended Freundlich) and mechanistic models (WHAM-VII) were used to predict metal solubility in the MA and Urban soils. The advantage of using E-values (ME) over metal extractable by dilute nitric acid (MExt) to represent the reactive metal fraction in predictions of solubility was very clear for WHAM-VII, but not for the ‘locally parameterised’ Freundlich model. This was almost certainly due to the strong links between pH and E-value becoming subsumed into the coefficient nominally describing the direct influence of pH on metal solubility in the extended Freundlich equation. However, overestimation of the solution concentration from WHAM-VII was observed for all five metals, and strongly correlated with soil pH. Fractionation information from WHAM suggested that the source of the model’s underestimation of metal binding most likely lay with errors in the description of metal binding by Fe and Mn oxides for Ni, Zn and Cd and humic acid (HA) for Cu. An additional factor is the absence of potential binding phases in the WHAM model, such as particulate CaCO3, and the greater diversity of active adsorbents in soils at high pH values. WHAM is based on the assumption that all metal bound to HA is labile. However, in the current study, ‘non-labile’ Cu, Zn and Pb fractions were observed in suspensions of HA extracted from grassland and peat soils. These were quantified by measuring metal E-values and EDTA-extraction of HA-bound metal using size-exclusion chromatography (SEC) coupled to ICP-MS to separate free and HA-complexed metal forms. Evidence of time-dependent metal fixation by HA was found for all three metals during the course of a 40 and 160 day incubation study. The proportion of non-labile Cu held by HA could be 40-50%. The presence of a non-labile metal fraction held by HA may substantially invalidate the assumption of reversible equilibrium which is central to all current geochemical models of metal binding to humic substances.

Climate change is a consequence of changing in climate on environment over the worldwide. The increase in developmental activities and Greenhouse Gases (GHGS) put a strain on environment, resulting in increased use of fuel resources. The consequence of such an emission to the atmosphere exacerbates climate pattern. There are numerous Climate Change Downscaling studies in coarse resolution, which have largely centred on employing the dynamic approaches, and in most of these investigations, the Regional Climate Model (RCM) has been reported to numerically predict the local climatic variables. The majority of previous investigations have failed to account for the spatial watershed scale, which could generate an average value of downscaled variables over the watershed scale. To address shortcomings of previous investigations, the work undertaken in this project has two main objectives. The study first aims to implement a spatially distributed Statistical Downscaling Model (SDSM) to downscale the predictands, and second to evaluate the impact of climate changes on the future discharge and peak flow. It is conducted based on the IPCC Scenarios A2 (Medium–High Emission scenario) and B2 (Medium–Low Emission scenario). The main objectives of the study are as follows: • To generate fine resolution climate change scenarios using Statistical Downscaling Model in the watershed scale, • To project the variability in temperature, precipitation and evaporation for the three time slices, 2020s (2010 to 2039), 2050s (2040 to 2069) and 2080s (2070 to 2099), based on A2 and B2 scenarios, • To calibrate and validate hydrological model using historical observed flow data to verify the performance of the hydrological model, • To evaluate the impact of climate changes on the future discharge and future peak flow for three timeslices: 2020s (2010 to 2039), 2050s (2040 to 2069) and 2080s (2070 to 2099). Thus, to meet the objectives of the study, projection of the future climate based on climate change scenarios from IPCC is carried out as the most important component in the research. The results of this research are presented as follows: • The study indicates that there will be an increase of mean monthly precipitation but with an intensified decrease in the number of consecutive wet-days and can be concluded as a possibility of more precipitation amount in fewer days. • The watershed is found to experience increased rainfall towards the end of the century. However, the analysis indicates that there will likely be a negative trend of mean precipitation in 2020s and with no difference in 2050s. The precipitation experiences a mean annual decrease by 7.9%, 0.6% in 2020s and 2050s and an increase by 12.4% in 2080s corresponding A2 scenario. • The maximum and minimum temperatures are likely to be increased toward the end of the century by 2.7oC and 0.8oC respectively when compared to the current observed temperature (1975-2001) at the Subang temperature station. • The average annual mean discharge is predicted to be decreasing by 9.4%, 4.9% and an increase of 3.4% for the A2 and a decrease of 17.3%, 13.6% and 5.1% for the B2 scenario, respectively in the 2020s, 2050s and 2080s. • The average annual maximum discharge is projected to decrease by 7.7% in 2020s and an increase by 4.2% and 29% in A2 scenario for 2050s and 2080s, respectively. But there will most likely be a decrease in the maximum discharge for all the future under B2 scenario. It is projected a decrease of 32.3%, 19.5% and 2.3% for 2020s, 2050s and 2080s, respectively. • The projected mean discharge indicates a decline in the months from January to April and also from July to August in all the three future periods for A2 and B2 scenarios. There is an increasing trend in the discharge of September and October in the 2020s according to the A2 and B2 scenarios. • The highest increase in precipitation frequency occurs in 2080s under A2 scenario in which the increase in the magnitude of 100 Return Year is found to be 88% greater than the one of the maximum observed. • The highest increase in flood frequency at Sulaiman streamflow station occurs in 2080s under A2 scenario. The increase in the magnitude of 100 Return Year is found to be 26.5% greater than the one of the maximum observed.

Dust from mining activities is produced from several unit operations and is often a serious problem to the industry, due to the influence it can have on human health and the safety record and productivity of a mine. So far, legislative parties and the industry have approached dust as an issue that needs to be controlled, only when a mining operation or process generates undesirable particulates. Nevertheless, new legislation and standards, such as the EU IPPC directive (Integrated Pollution Prevention and Control) and air quality strategies aim to drive mining companies to incorporate dust assessment planning that will be implemented through the whole life cycle of the mine. Mitigation and monitoring practices as well as health surveillance programs will need to be clearly defined. This project’s purpose is to understand how mining processes and in particular how the mechanisms inherent within common unit operations (i.e transfer processes using haulage roads or conveyor belts, the tipping, loading and stockpiling process, the screening process etc) result in the generation of dust. If the operation of unit operations could by optimized to produce less dust, then a “fit-for-purpose” strategy for dust minimisation could be developed to follow exploitation, processing and production demand. The literature on dust from mining operations identified that generation of fines/dust occurs due to the presence of the mechanisms of abrasion and impact. Based on this logic, an experimental methodology was developed, which aimed to assess how dust was generated for each different mechanism and for a variety of ores of different mineralogy. Five different ores were tested, a limestone, talc, an iron ore, a lamproite and a copper ore and the same experimental methodology was followed for each. Experimentation using the HSE-WSL tumbling mill test determined that under the effect of abrasion, ores yielded higher dustiness values during longer tumbling times, whilst parameters such as the sample mass and the particle size distribution of the feed sample could also influence the dust generation patterns. The findings of the computational modelling (discreet element modelling) and experimentation (high speed video recording) suggested that control and optimisation of operational parameters (e.g mill velocity, or tumbling time) within processes that involve abrasion, such as the use of conveyor belts, mills, and screens could minimise the potential of dust generation by this mechanism. The use of a novel impact test determined a positive relationship between the energy input and the particle size distributions of the broken particles, as well as the accumulation of fine particulates in the range of dust (<75μm). Also an increase in the bulk volume of ores resulted in larger quantities of fine particulates. These observations suggest that it is possible to reduce dust in processes that involve drop from heights and impaction (i.e transfer points in conveyor belts, tipping, loading) by adjusting the energy input and the bulk volume of ore at impact to as low a level possible. Particle size analysis of the produced dust fractions were found to be material dependent and varied considerably for the different ores. Almost all materials produced significant amounts of ultra fine particles below 10μm and 2.5μm, both under impact and abrasion, which reveals the potential adverse impacts to the environment and human health. Quantitative mineralogical analysis using the mineral liberation analyser determined that the dust fraction presents a different composition to that of the ore. Comparison of the results collected for the five different ores using the HSE-WSL mill and the impact test identified that certain materials yielded high dust levels under abrasion and low under impact. Therefore it would be expected that dust control approaches for such materials would differ according to the mechanisms of the involved process and the mineralogy of the sample. According to the findings of this study a reduction in dust produced from mining unit operations could be possible by optimising the involved processes either by altering their operating parameters (drop height during tipping, velocity of conveyor belt) or by optimising the design of processes so as to reduce abrasion or impact. New legislation such as the EU IPPC directive has already started considering such an approach as important, and newly developed Best Available Techniques documents refer to this as the primary step companies should follow to minimise dust. Additional advantages of this approach are that it can reduce cost for dust control by making use of less conventional mitigation practices, and in the long term it could also minimise the utilization of energy and water going to suppression, extraction and dust collection systems. In certain cases the proposed route could also optimise the production chain, especially where the generation of fines is undesirable (e.g iron ore processing or aggregates production).

Results are presented from sets of field and laboratory experiments conducted to measure and quantify the Hydraulic Residence Time Distribution in treatment ponds containing vegetation. The field measurements were taken in the Lyby field pond (Sweden) with complementary experiments on a distorted, laboratory scale model pond designed and built in the University of Warwick’s engineering laboratory. Rhodamine WT Dye tracer experiments were used in both the Lyby field pond and the distorted physical scale model to investigate vegetation and discharge affects on HRTD characteristics and the technique of PIV (Particle Image Velocimetry) was used in the distorted physical scale model to investigate how surface flow profiles were affected by different vegetation and discharge configurations. The results show that the distorted physical scale pond did not reflect the HRTD characteristics of the field site, with the actual residence time, (tm), for the distorted physical scale pond ranging from 85 % to 125% of its nominal residence time. For the distorted scale model, pond vegetation and discharge did not affect the relative HRTD centroid, em, or the actual residence time, tm. This finding is attributed to the unique pond geography and associated aspect ratios However, flow rates did have a significant effect on the HRTD e0 (time of first dye arrival at the outlet) and ep (time of peak dye concentration). Changes in vegetation were found to have little effect on e0 and ep. For the laboratory pond, vegetation had a significant control on the surface flow field whereas, flow rates did not – the latter suggests that surface flow fields are not representative of the internal flow field in different layers of the pond. The experiments demonstrate that the specific shape of the distorted physical scale pond in this study enables optimal actual resident times to be achieved over a wide range of vegetation and flow rate configurations. If full scale field ponds based upon this design give the same stable centroid results, then this would be a substantial breakthrough in pond design, which would aid the design and management of pond treatment and allow more robust optimisation of treatment efficiency.

The performance and sustainability of ground source heat systems is dependent on the thermal behaviour of neighbouring ground. This thesis describes a comprehensive experimental and numerical investigation into the ground behaviour in response to horizontal ground source heat systems. Experimental investigations comprised of a field-scale monitoring scheme, designed and implemented in a horizontal ground source heat system providing space heating to a domestic property located in Mid-Wales, UK. A high resolution ground temperature data-set has been compiled over a 13 month period via 112 thermistors buried in the ground. Further data-sets representing the climatic variables and heat pump behaviour were also compiled over the same period, facilitating a thorough investigation of the ground behaviour in response to heat extraction at the site. Soil properties were also measured at the site as part of a larger site investigation undertaken. The numerical model applied is a coupled thermal-hydraulic (TH) model previously developed at the Geoenvironmental Research Centre (GRC). The GRC’s current model was extended to include developed boundary conditions for the TH simulation of horizontal ground source heat systems, describing the soil’s interaction with the ground-loop and atmosphere. Developments were subject to rigorous validation including comparisons with ground-data collected at the experimental site. The validated model was applied to investigate the long-term ground behaviour at the monitoring site and the effects of different surface materials on the ground behaviour including the recharge process. Finally, the model was applied to investigate the validity of an existing design code for ground source heat system design. The resolution and duration of the collated data-set facilitated extensive analysis, including a thorough investigation of the ground thermal distributions resulting from heat extraction and recharge. Findings indicated unsymmetrical distributions, highlighting potential avenues for system optimisation. Further to this it can be said that the data-set, in its own right, is a significant contribution to the scientific community and is able to provide a means of validation for future models. Results from the numerical investigation indicated that the ground thermal behaviour exhibits an annual cyclic pattern after approximately 3 years. From a holistic perspective, the results demonstrate that horizontal ground source heat systems can provide a sustainable means of providing space heating. Further long-term studies investigating the effects of surface materials show that more thermal energy can be sustainably extracted from systems with urban surface types (i.e. asphalt and brick). Investigations into an existing design code revealed that a ground-loop designed to meet the loads of the monitoring site resulted in unsustainable heat extraction, in doing so demonstrating the simplification of ground-loop burial depth within the current design process. Knowledge regarding the ground response to horizontal ground source heat systems has been furthered through the collection and analysis of field-scale data within this study. Further to this, the boundary developments and validation undertaken have allowed for a more thorough investigation of the long-term ground behaviour than previous studies.

Several methods of river water quality assessment techniques have been introduced. Among the most commonly used are the water quality index system and classification scheme. These two systems are designed to simplify the huge amount of water quality data down to its simplest form, while retaining the essential meaning of the information. They offer the means for measuring the effectiveness of pollution abatement strategies by comparing the status of water quality both temporally and spatially. In this way, it is useful for management purposes, especially in determining priorities for resource allocation and planning of new development areas. The water quality index system and the classification schemes currently available, however have some limitations in their structural design. They often exhibit inherent loss of information, are complex and may involve subjective judgement in their interpretation. However, because of the critical issues on water pollution and the scarcity of water resources, these systems are being applied despite of these limitations. The current situation is that, different countries are applying different models of water quality assessment system. Based on the limitations of the existing assessment systems, it is appropriate to explore other approaches that can be more flexible, robust to noisy data, and adaptable to new form of environmental data, in order to provide direct and prompt results for classifying of river water quality. One avenue for research is that based on Artificial Neural Network (ANN). Artificial Neural Network comprises of several techniques. One of this technique that is widely being used is the Back-Error Propagation (BEP). BEP of ANN was used in this research in conjunction with the Interim National Water Quality Standard (INWQS) data for Malaysia. The findings of the study shows that the classification results based on the evaluation of the water quality variables were good when compared with the results obtained from other water quality classification models, which include: the Department of Environment Water Quality Index (DOE-WQI), the Harkins'-WQI, Mahalanobis Distance Classifier, Maximum Likelihood Distance Classifier and the Decision Tree Classifier. The accuracy for BEP of ANN was found to be 86.9% and correlated well with all of these five models. The highest correlation was, with the Mahalanobis Distance Classifier and the DOE-WQI. The analysis on sensitivity shows that the BEP of ANN was sensitive to Dissolved Oxygen, a condition similar to the DOE-WQI model. Comparisons were made with two types of BEP of ANN architecture, a simple network with less number of hidden nodes and a relatively complex network with more hidden nodes. It can be concluded from the analysis that a small and simple network performed well with large samples and with test data that are widely distributed than the complex network with more hidden nodes. Using the same model, different approaches were used in evaluating the classification of water quality were applied, such as the used of the land use variables and hydrological features (LUVHF) to replace the water quality data. Using these variables, the performance of the BEP of ANN in classification of water quality was low (24% and 31%). However, its performance can be improved, if more samples with wider range of LUVHF were used. Throughout this study, the BEP of ANN model has shown some remarkable achievements. In view of these, several knowledge contributions have been made. The first contribution is the flexibility of the system approach and operationally simple to perform. Secondly, it provides a practical approach in classification of river water quality, such that through a single network computation of a sample, the results are presented promptly as the probability value and the class grade value. The third contribution is that the water quality can also be classified based on the land use variables and hydrological features, without dependence on water quality data. This approach is suitable for remote areas, where accessibility is relatively difficult.