The Indonesian natural rubber industry is facing a reduction of natural rubber supply because many rubber plantation areas are entering their low-productive phase. There are significant opportunities to sustain future supplies by managing the replanting of these low-productive areas. Early discussions with Indonesian stakeholders in Northern Sumatera indicated a demand for strategic planning tools to support decisions related to replanting and described significant effects of replanting to the whole region and individual players within the network. The examples of these effects include the level of future supply from the region and the flow of natural rubber between players, which in turn affects income of these players. In addition, a key finding from a review of literature indicated the importance of integrating three dimensions of sustainability (economic, environmental and social) in planning for future supply networks. The aim of this research was to explore ways in which social, economic and environmental aspects of sustainability can be used to inform decisions related to the formulation of replanting policies. This research contributes a sustainability assessment and optimisation method for use in the formulation of sustainable replanting policies. The sustainability assessment method consists of suitable sustainability indicators and an integrated assessment tool. Indicators at district and individual levels for three dimensions of sustainability were identified through case study investigation and a review of literature. A combination of system dynamics and agent-based simulation was developed as the integrated assessment tool, which demonstrates an ability to assess long-term sustainability impacts of replanting at district and individual levels. The optimisation method was developed to enable the used of the integrated assessment tool to inform the formulation of replanting policies, comprising a combination of composite indicators and dynamic programming. Composite indicators were used to translate sustainability impacts of replanting calculated using the integrated assessment tool into indices of replanting impacts and dynamic programming was used to determine optimum replanting policies using these indices. The application of the method using real world data demonstrated the process of formulating sustainable replanting policies for the North Sumatera Natural rubber industry.

The need to reduce the cost of components is driving more and more machine elements to operate under mixed lubrication conditions. With higher operating pressures, the lubricant film is becoming thinner and eventually reaches nanometre scales, comparable to the surface roughness. Thus, understanding the mixed lubrication phenomenon is becoming increasingly important. However, the mixed lubrication phenomenon is difficult to capture experimentally and the lubricant additive ZDDP (Zinc Dialkyl Dithio Phosphate) shows its full antiwear character in the mixed lubrication conditions. This research stems from the need for models that can simulate contact mechanics, lubrication and tribochemistry in a single framework. This is the key to understanding and optimizing the lubrication systems to meet future needs. To this end, a numerically efficient procedure based upon the tridiagonal solution of the Reynolds equation is developed and is implemented, in Fortran to solve the equations line by line to incorporate more information from the current iteration step. The asperity contacts are handled by the unified solution algorithm. A new strategy to simulate plastic deformation in a lubricated contact is developed. Under practical loading conditions, the pressures inside the contact can reach values far above the material yielding limit. Thus, an efficient numerical scheme is devised to include the elastic perfectly plastic behaviour in the EHL solution procedure to simulate realistic contact conditions with minimal increase in computational cost. The Boussinesq deformation integrals result in a convolution of pressure and the deformation which is solved using Fast Fourier Transforms (FFTs) by modifying the solution domain to create a cyclic convolution. Code is developed to allow exploration of the highly optimized C-based library (www.fftw.org). The use of FFTs speeds up the solution process many times and makes the use of denser grids and larger time scales accessible. A mesh size of 129 x 129 is found to give reasonable results. The simulation results from the current study agree very well with the previously published results. The evolution of contact area ratio and the central film thickness exhibit a Stribeck type behaviour and the transition speeds at which the contact transits from EHL to mixed and from mixed to complete boundary lubrication can be precisely identified. Existing tribofilm growth models developed for boundary lubrication are reviewed and a model based on the interface thermodynamics is adapted and integrated with the mixed lubrication model to simulate tribochemistry. The problems with existing EHL concepts such as lambda ratio and central film thickness are identified and new definitions are proposed to allow understanding of the mixed lubrication mechanics. The mutual interaction between the tribofilm growth and lubricant film formation is studied. Finally the wear of the tribological system is studied and the wear track profiles are predicted. The new model is then applied to study a ball-on-disc system to explore wear, tribochemistry and roughness evolution. The ZDDP tribofilm growth is studied and the it is found that the final ZDDP tribofilm thickness is very weakly affected by increasing SRR but the rate of formation and removal are strongly affected by the SRR value. The tribofilm growth results are validated against published numerical and experimental results. It is found that the antiwear action of the ZDDP tribofilm is not only due to its chemical action but the ZDDP tribofilm helps to entrain more lubricant and improves contact performance. The presence of tribofilm roughens the contact and the contact area and load ratio both increase due to tribofilm growth. It was also found that the use of conventional EHL parameters to analyse the behaviour of tribosystem is misleading. The flattening of the roughness inside the contact and the proper identification of the central film thickness are crucial to the interpretation of the mixed lubrication results. The roughness of the ball generally decreases due to wear but the presence of tribofilm limits this reduction. Contrary to this, the surface roughness of the ball generally increases due to wear but the presence of tribofilm results in increased roughness of the ball.

In the future an automatic robot fabrication system is envisioned, that would enable a non-specialist user to parametrically design, and autonomously manufacture a bespoke robot, without the need for manual intervention or assembly. Additive Manufacturing (AM) is an attractive method for this as it is digitally driven and tooling free, which increases the speed and flexibility of production, while reducing costs for low volumes. Another unique capability of AM is the ability to embed components within the structure of a robot, while it is being manufactured. However, the state of the art embedding methods involve a series of manual steps. Additive manufacturing also has some inherent limitations which arise from building up parts from many discrete, planar layers. These include; anisotropic mechanical properties; curves approximated by discontinuous steps; and overhangs requiring support. Addressing these issues, while also enabling functional mechatronic components to be embedded, requires a new approach to Additive Manufacturing. This work introduces a novel 12-axis Additive Robot Manufacturing System (ARMS). This is shown to successfully 3D print high quality parts, comparable to other Fused Filament Fabrication (FFF) systems, using industrial robot arms. Using the multiaxis (i.e. > 3 axes) capabilities, it is demonstrated that the orientation of a print with respect to gravity has no effect upon its surface quality, but the relative orientation of the geometry to the layers has a significant influence. Using this relationship, components are printed using a dynamically varying build orientation, enabling unsupported, 90° overhangs with a constant, tuneable roughness to be produced. ARMS is then used to print nonplanar layers with greater curvature than has previously been demonstrated. Conformal layers are shown to improve the strength of curved parts by 57%. The capabilities are combined to manufacture the first mechatronic system with an integrated actuator that required no manual assembly, intervention, or post processing during or following the AM process. It is concluded that the multiaxis AM system successfully improves upon conventional planar deposition by overcoming key drawbacks to FFF, and could be a key enabler to a new AM process for the manufacture of integrated mechatronic and robotic devices.

Challenging times lie ahead for commercial aerospace, facing regulatory pressure to reduce emissions on one side and the potential of increased competition on the other, a continuation of the business and engineering philosophies which led to such a healthy orderbook in the past, cannot be guaranteed for the future – substantial, disruptive change is required. Additive Manufacturing (AM) and Topology Optimization (TO) are two technologies under investigation by Airbus and others which have promised to deliver such change. Problematically, both are expert level technologies with enormous complexities and thus their application is commonly applied only where justification of such skills for such lengths of time can be considered to be economically viable. However, whilst there are indeed gains to be had in such large, complex structures, their numbers on commercial aircraft are few. Conversely, there are literally thousands of small, heavy, metallic components which would benefit from the application of these technologies if the cost of technology application could be reduced. The aim of this research is to deskill the application of TO and AM by automating the process of TO specific to manufacturing via AM and thus reduce the cost of its implementation and increase the practicality of its application. Through a survey of the Airbus user community, a standardised series of tools, inputs, outputs and process was developed, culminating in an analysis of time consumed during a series of optimization tasks. From this list of tasks and the time lost to each, a series of targets for automation were identified and researched. Using a series of interconnected codes and scripts, pre-processing phases such as design space creation, meshing and loading application were automated and applied to a common FEM template. Within this template, generic material and geometric capability figures for AM Ti64 Grade 5 were established via bespoke testing on a range of AM platforms under common parameters and builds. After this, methods for automated design extraction back to parametric CAD were investigated and performed, establishing a direct link between the FEM and the output CAD to enable rapid design development. The combined series of automation steps leads to an almost 75% reduction in total non-recurring cost for optimization and design of small components. Whilst not, as yet, wholly industrialised and implemented within Airbus, research from the early phases is now in use for MDO tools within Airbus and Airbus Group.

A generalized scattered noise behavioral model for time-invariant nonlinear microwave circuits is presented. The formalism uses noise waves and large-signal scattering functions known as X-parameters to extract a multi-port network's noise correlation matrix. Further processing yields figures-of-merit including effective input noise temperature and noise factor. Within the small-input signal space, it will be shown that the above expressions reduce to a familiar form describing noise wave influence governed by the network's S-parameter functions. Using the generalized form, two examples given in context to embedded nonlinear one-port and two-port configurations are offered with each presented to matched termination networking. Both cases use a passive source and load in the analysis. Numerical versus simulated experimental results will be compared. Results in the two-port case yield its noise factor. Lastly, pursuant to this study, experimental work involving software simulation and hardware measurement activities will be proposed.

This thesis describes the development of on-chip terahertz (THz) waveguides for use in ultrafast magnetic measurements. An epitaxial transfer method is used to bond photoconductive switch material to quartz substrates. This, combined with optical lithography methods, allows on-chip generation and detection of THz pulses. Normally, the metal in a waveguide has a high conductivity (eg. gold or copper) in order to reduce ohmic loss. However, if this metal is instead a magnetic multilayer, then the THz pulse interaction with the metal can be used to measure ultrafast magnetic effects. This was performed using Cobalt/Copper multilayers, and a THz GMR was seen in a transmitted pulse. Additionally, an ’inverse’ GMR was noted in input pulses and in the pulse tail of transmitted pulses. It is believed that this is the first time such behaviour has been documented. Lateral spin valves offer the ability to separate charge and spin effects; however, the dimensions of these devices is in the hundreds of nanometers range. Coplanar waveguides were used to confine and concentrate THz pulses into this spatial regime. Curved waveguides were fabricated to prove confinement, and magnetoresistance properties were used to demonstrate field concentration. Devices were then designed to allow picosecond current injection into a lateral spin valve geometry. Initial results are obscured due to interference from the excitation pulse, but improvements to the design and fabrication could pave the way for this technology in the future.

In recent years there has been an increase in the number of cars using automatic transmissions and in these systems the control units are often placed in contact with the Automatic Transmission Fluid (ATF). It is therefore important that the additives present in the ATF do not adversely interact with the transmission components, particularly those which are copper-based, such as the solenoids. There is currently very little literature on copper corrosion in oil based systems particularly when looking specifically at the interaction between ATFs and copper. This study looks at the interaction of some common ATF additives with copper surfaces. This is achieved through the combination of simple immersion tests, conducted on coupons, and resistance tests, carried out on wires. The coupons allow detailed analysis of the surface by FTIR, SEM and XPS. The wire tests monitor the resistance of a thin copper wire which can be directly related to the radius and indicate when there is loss of material. This allows the corrosion of the wire to be monitored in situ which is not achievable through immersion tests. The study looked at the standard ASTM D130 rating method as a basis for attributing corrosion to coupons after they had been immersed in a test fluid. It was found that this rating did not really correlate with any other measures of corrosion such as weight change, amount of copper in the used test fluid or radius change measured using the copper wire resistance test. It was concluded that whilst this test may be suitable for screening large numbers of fluids it provides little information on the actual corrosion taking place on the surface and in some instances film formation can be identified as corrosion. The study showed that increasing the temperature at which testing took place did not always speed up the corrosion process. In one instance increasing the temperature slowed the rate of reaction but in most cases increasing the temperature changed the mechanism of corrosion. The most extreme case of this was with the thiadiazole-based corrosion inhibitor which showed good inhibition until 150 °C when it broke down and caused pitting on the copper surface. The wire tests were able to show which additives protected the copper from corrosion, which gave constant corrosion, and the effect of combining additives. An antagonism was seen at 120 °C between dispersant 1 and a mix of antiwear, antioxidant and friction modifier. Detergent 2 combined with a mix of antiwear, antioxidant and friction modifier at 150 °C was also antagonistic. SEM images of the surfaces of coupons tested in full formulation fluids were able to be compared with those tested in single additives and simple mixtures to give information on which of the additives were causing the observed surface structure. This worked well for formulations where the additives were at similar levels to those tested individually but not for those where the concentration of additives was much lower. For each of the individual additives tested a mechanism of interaction with the copper surface has also been proposed.

No description available.

Evolution of wireless communication systems towards high flexibility, low cost and high efficiency leads to tremendous activity in the area of microwave filters. In an RF front-end of a cellular radio base station, signals are being transmitted and received simultaneously. In the receive band, there are chances of intermodulation products from the power amplifier being fed to the receiver,thus the transmit filter must have a very high level of signal rejection. Furthermore, the transmit filter must also have low passband insertion loss since it impacts the power transmitted and the overall transmit system efficiency. Recently, filters with dual-mode operation were being investigated due to their ability to produce two degenerate modes using a single physical structure; therefore, the size and cost of the filter can be reduced without compromising any figure-of-merits. A dual mode suspended substrate stripline filter is presented in this thesis. These filters enable achieving low insertion loss, high Q, high selectivity and good spurious response. Initially, a dual mode ring resonator structure is investigated using suspended substrate stripline technology. This technology is used due to its advantages which are comparable to microstrip or any other planar transmission lines. The HFSS three dimensional finite element method (FEM) is used to evaluate the resonant frequency, quality factor and the first harmonics. A second order suspended substrate stripline filter was designed with capacitive coupled input and output feeding method. The input and output feed were positioned 90 degree from each other while a notch was used in this filter to couple two degenerate modes which also control the bandwidth of the filter. A high performance Generalized Chebychev filter was designed to meet the stringent electrical requirement in the RF front-end of a cellular radio base station. With this fourth order filter, four finite frequency transmission zeros were achieved due to phase cancellation between two paths which results in high selectivity filter response. Metal tuning screws were added to improve any practical imperfections. Finally an asymmetrical Generalized Chebychev filter was designed with real frequency transmission zeros positioned on one side of the passband. With this design, the aim of achieving higher selectivity filter response above the passband was demonstrated.

Data centres are an increasingly large contributing factor to the consumption of electricity globally, and any improvements to their effectiveness are important in minimising their effect on the environment. This study aims to achieve this by looking at ways of understanding and more effectively utilising IT in data centre spaces. This was achieved through the testing of a range of ways of creating virtual load, and employing them on servers in a controlled thermal environment. A Generic Server Wind-tunnel was designed and built which afforded control of thermal environment and six different servers were tested within, yielding results on performance and thermal effect. Further testing was also conducted on a High Performance Computing server with a view to understanding the effect of internal temperature on performance. Transfer functions were created for each of the six servers, predicting behaviour reliability for five output functions and validating the developed methodology to an appropriate accuracy. The trends seen and the methodology presented should allowed data centre managers better insight into the behaviour of their servers.