VIPS is an open source image processing library which has evolved from many EU projects on imaging. It is designed for speed, automatic multithreading and ease of programming. Its NIP GUI provides a spreadsheet-like interface with its own functional language for scripting. It has a wide develop/user community around the world and runs on almost any platform (using C/C++/Python). It has been used for mosaicing images, dealing with multi-GB images and colour calibration.
The Glacsweb designed a sensor network to monitor glaciers - including producing new hardware (node) designs, new radio communication network protocols and robust control and telemetry. It deployed systems in Norway and Iceland which produced continuous data previously unseen by glaciologists. The interdisciplinary research involved computer science, electronics, mechanical engineering, glaciology and GIS.
This project has been established by the Electrical Power Engineering Group at the University of Southampton to address some fundamental issues associated with the application of superconducting technology to power apparatus. One of the critical aspects for the application of high temperature superconductor (HTS) technology is suitable electrical insulation materials. In many HTS devices, liquid nitrogen (LN2) is a cheap and excellent cooling medium and also a good electrical insulator. However, bubbles are easily formed due to trapped vapour on surfaces during filling, thermally or electrically induced. These bubbles are known to reduce the breakdown withstand level.
This project is studying the thermally induced bubble dynamics and boiling process under electric fields. To date results have been obtained using rod-plane and inclined-plane electrodes. These results are useful for understanding electrohydrodynamic (EHD) phenomena influencing the bubble behaviour in liquid nitrogen and for developing a means of removing the gas from the cryostat of HTS transformers or fault current limiters (FCL) during ââ¬Åquenchââ¬?. In addition, the EHD technique can be used as an effective heat transfer enhancement method in liquids. Obtained experimental results of liquid nitrogen boiling curves under applied electric fields will allow future cooling equipment to exploit this phenomena.
To develop algorithms and protocols that can improve the scalability of sensor networks by using cluster-based routing mechanisms. The network must be able to balance energy consumption between nodes and maintain a low message latency.
A partial discharge (PD) within a transformer can be considered as a high frequency transient event, therefore the development of high frequency models may inform the process of pd source location and evaluation. PD activity in large transformers can be detected using wideband sensors (vhf up to 500MHz)and therefore any model should be broadband. The use of a laboratory experimental model containing both interleaved and plain disk windings has provided a range of time and frequency response data that has yielded models based on lumped circuit parameters. Other modelling approaches are being investigated as well as techniques for determining PD source location based on the developed models and measurements of PD activity made at the bushing tap and neutral to earth points.
The insulation used for a high voltage application is one of the most important components in any design because breakdown of the insulation may cause total failure of the whole component. The electrical insulation system is subjected to high electrical, mechanical and thermal stresses which can age and degrade the insulation to the point where partial discharges (PD) can become a regular occurrence ultimately leading to total breakdown. Detection and analysis of PD activity is therefore important to ensure the health and lifetime of any high voltage asset.
Partial discharge is a discharge event that does not bridge the electrodes within an electrical insulation system under high voltage stress. In high voltage components, the measurement of partial discharge is used in the performance assessment of an insulation system. Through modelling and measuring the discharge process a better understanding of the phenomena may be attained. This project describes the development of a mathematical model describing partial discharge in a spherical cavity and ellipsoidal cavity within a homogeneous dielectric. The model developed is used to study the influence of the applied frequency on partial discharge activity and also the influence of the cavity size and cavity location within the insulation material on partial discharge frequency dependent activity. The simulation results can then be used within insulation diagnostics to assist in the assessment of the performance of the insulation system.
A number of technical and policy issues are of concern within the University around e-assessment, including:
The University has received funding to accelerate the process of both implementing an open source, service based solution to institutional e-assessment and addressing institutional change by engaging academics and students in co-design and co-deployment. In line with the recently introduced University e-Learning Enhancement Strategy, which places the quality of student learning as its first objective, the EASiHE project intends to provide an open source solution for formative assessment by integrating services currently available within the JISC eFramework.
The project will:
The main deliverables include:
As partners in the JISC Institutional Innovation Support Project, EASiHE will be actively involved in producing briefing materials and training activities for exemplary practice.
The aim of this research is to beneficially redistribute energy consumption in wireless sensor networks. Typically, sensor nodes have scarce energy resources and must rely on energy harvesting to prolong their life. By contrast, the sink node is often integrated into a higher-level system and can benefit from the accompanying plentiful energy resources, such as the mains. It is therefore desirable to redistribute energy consumption from the sensor nodes to the sink node. This can be achieved using iterative decoding.
This project is investigating the development and demonstration of a pervasive home welfare monitoring system. The system aims to use unobtrusive sensors for early detection and automated reporting of deteriorating physiological health parameters and enabling independent living.
To develop algorithms and protocols that can improve the scalability of sensor networks by using cluster-based routing mechanisms. The network must be able to balance energy consumption between nodes and maintain message latency under predetermined values.