School of Electrical & Information Engineering (ETDs)

Permanent URI for this communityhttps://hdl.handle.net/10539/37968

Browse

Now showing 1 - 6 of 6

Action volume ratio: a method to classify the danger of lightning in any given volume
(University of the Witwatersrand, Johannesburg, 2015-05-07) Dickson, Andrew Stuart
The annual number of injuries and fatalities from lightning has, over the last century, been on a steady decline. This is primarily due to urbanisation and movement away from agriculturally intensive activities. In countries with a high urban population, the incidence of lightning fatalities is below 1 death per million people per year. However, in countries with a larger rural population, this rate is significantly higher, ranging between 8 and 15 deaths per million people per year. There has been a large drive towards educating the general public about the dangers of lightning and methods to avoid being in a dangerous situation. However, fatal lightning events still occur on a regular basis. There are currently no methods to determine the risk of lightning to living beings in open spaces. The international standard (IEC 62305-2) provides a method for the assessment of risk to living beings within a structure, and up to three metres outside of it. Considering that the majority of deaths by lightning occur outdoors, a method of determining these risks is necessary. The Action Volume Ratio (AVR) is proposed as a new method for the analysis of the danger of lightning in any volume. It considers the dangers of all lightning injury mechanisms in relation to the objects in the space, which are assumed to be the preferential points of strike. A union of the dangerous volumes is then formed, and a ratio to the total volume is created. The AVR uses accepted electrical engineering equations to determine the dangerous areas, and places no reliance on probability theory, which can, in many cases, skew the results of a lightning risk analysis process. The AVR can be combined with lightning ground flash density data to indicate the incidence and frequency of dangerous events within a given volume.
Discrimination between nearby and direct lightning strikes to a long operational medium voltage line to assist in the determination of the basic insulation level (BIL)
(University of the Witwatersrand, Johannesburg, 2022) Van Schalkwyk, Willem Jacobus Dirkse; Gomes, Chandima; Van Coller, John
The lightning performance of a Medium Voltage (MV) line needs to be divided into two categories: lightning performance due to nearby lightning and direct lightning strikes. A better nearby lightning performance requires a higher Basic Insulation Level (BIL) while the direct lightning performance requires a lower BIL to minimize equipment failure. The electromagnetic coupling models for calculating the Lightning Induced Overvoltage (LIOV) on a line are complicated and reliant on accurate input data. Therefore, short floating lines (< 3 km) were used to calculate the line’s lightning performance. The models were then verified with LIOV measurements on these short floating lines and the results were normalized to estimate the lightning performance of long operational lines. These estimations for long lines could never be verified due to the complexity of calculations and the cost and logistics of equipment to do measurements on long operational lines. A new methodology to measure the lightning performance of a long operational line was developed and verified. The obstacles such as the long line length, the changing soil resistivity, the lightning channel properties and the rapid attenuation of the LIOV along the energized line have been used as an advantage in the new method. The finite soil conductivity was used as an advantage to distinguish between nearby lightning and direct lightning strikes while the power frequency current was used to determine whether the LIOV exceeded the line BIL and caused a line fault. An existing electromagnetic coupling computer model, the ATP-EMTP, was verified with the existing ERM using the same input data. The ATP-EMTP model was then extended to include the equipped long MV line. There was no existing model or measurements to which the results could be compared to. Actual nearby lightning and direct lightning strikes of which the termination point was known were used to verify the new methodology. A significant size database (consisting of 38 675 lightning flashes and 1 155 line faults over two years) was used to evaluate the IEEE Std 1410-2010 estimated lightning performance of the long line.
Evaluation and algorithmic adaptation of brain state control through audio entertainment
(University of the Witwatersrand, Johannesburg, 2023-12) Cassim, Muhammed Rashaad; Rubin, David; Pantanowitz, Adam
This dissertation presents the design and evaluation of a system that can alter the dominant brain state of participants through audio entrainment. The ‘rch broadly aimed to identify the possible improvements of a dynamic entrainment stimulus when compared to a set entrainment stimulus. The dynamic entrainment stimulus was controlled by a Q-Learning (QL) model. The experiment sought to build on previous research by implementing existing entrainment methods in Virtual Reality and dynamically optimising the entrainment stimulus. The neurological effects of the stimuli were evaluated by analysing electroencephalogram measurements. It was found that a set 24 Hz entrainment stimulus increased the power of Beta band brain waves relative to a control condition. Further, contrary to existing research, it was found that the entrainment stimulus did not have a notable effect on brainwave connectivity at the entrainment frequency. The study subsequently evaluated if the QL agent could learn to optimise the entrainment stimulus. The agent was allowed to switch between an 18 and 24 Hz entrainment stimulus and succeeded in learning an optimised policy. The QL driven stimulus yielded results that generally exhibited the same characteristics as the set entrainment stimulus when using power and connectivity analysis methods. Furthermore, the power analysis indicated that the QL driven stimulus was able to affect a broader range of frequencies within the targeted band. The QL driven stimulus, additionally, resulted in higher meta-analysis metric values in some aspects. These factors indicate that it was able to have a more consistent impact on targeted brain waves. Lastly, results from participants whose stimulus was controlled by a QL driven stimulus using optimal actions indicated that the optimised actions created a more sustained increase in Beta band activity when compared to any other results, indicating the impact of the optimised policy learned.
Model Propagation for High-Parallelism in Data Compression
(University of the Witwatersrand, Johannesburg, 2023-10) Lin, Shaw Chian; Cheng, Ling
Recent data compression research focuses on the parallelisation of existing algorithms (LZ77, BZIP2 etc.) by exploiting their inherent parallelism. Little work has been performed on parallelising highly sequential algorithms, whose slow compression speeds would benefit the most from parallelism. This dissertation presents a generalised parallelisation approach that can be potentially adopted by any compression algorithms, with model sequentiality in mind. The scheme presents a novel divide-and-conquer approach when dividing the data stream into smaller data blocks for parallelisation. The scheme, branching propagation, is implemented with prediction by partial matching (PPM), an algorithm of the statistical-modelling family known for their serial nature, which is shown to suffer from compression ratio increases when parallelised. A speedup of 5.2-7x has been achieved at 16 threads, with at most a 6.5% increase in size relative to serial performance, while the conventional approach showed up to a 7.5x speedup with an 8.0% increase. The branching propagation approach has been shown to offer better compression ratios over conventional approaches with increasing parallelism (a difference of 11% increase at 256 threads), albeit at slightly slower speeds. To quantify the speedup over ratio penalty, an alternate metric called speedup-to-ratio increase (SRI) is used. This shows that when serial dependency is maintained, branching propagation is superior in standard configurations, which offers substantial speed while minimising the compression ratio penalty relative to the speedup. However, at lower serial dependency, the conventional approach is generally preferable, with 9-16x speedup per 1% increase in compression ratio at maximal speed compared to branching propagation’s 6-13x speedup per 1%.
Peak-to-average power ratio reduction in optical-OFDM systems using lexicographical permutations
(University of the Witwatersrand, Johannesburg, 2023) Niwareeba, Roland; Cox, Mitchell A.; Cheng, Ling
The work presented in this thesis extends and contributes to the research in reducing the high Peak-to-Average Power Ratio (PAPR) in optical-Orthogonal Frequency Division Multiplexing (OFDM) systems using probabilistic-based and hybrid techniques. Whereas the high PAPR problem has been extensively studied and a number of solutions provided for the conventional Radio Frequency (RF)-OFDM systems, there are only a few solutions proposed specifically for PAPR reduction in optical-OFDM systems. Although the probabilistic-based techniques such as Conventional Selected Mapping (CSLM) and Data Position Permutation (DPP) result into significant PAPR reduction performance with negligible Bit Error Rate (BER) degradation, the resulting increase in both hardware and computational complexity as a result of a large number of Inverse Fast Fourier Transform (IFFT) operations that have to be performed to generate the candidate signals is still a major drawback. In order to reduce the complexity, in this research, two techniques which are applied in opticalOFDM systems are proposed. The first technique is the hybrid method composed of a modified CSLM and µ-law companding techniques called Low Complexity Hybrid Selected Mapping (LCHSLM). The proposed method achieves almost 50% reduction in complexity compared to CSLM with less BER degradation. The second technique based on lexicographical permutations called Lexicographical Symbol Position Permutation (LSPP) works by dividing the optical-OFDM symbol into a number of sub-blocks and performing lexicographical permutations to obtain the candidate signals after the IFFT operations. In the proposed LSPP, all the candidate permutation sequences are not obtained at once unlike in the DPP where the number of candidate permutation sequences increases at a factorial rate of growth as the number of sub-blocks increases resulting in a more complex system. Additionally, the research proposes an algorithm where a threshold PAPR value is introduced and the candidate signals are generated until a candidate with a PAPR value less or equal to the threshold is obtained. The results show that the complexity in terms of IFFT operations can be reduced substantially depending on the selected threshold and the number of candidate signals. Furthermore, the research introduces a new algorithm based on the global gain (net gain) to determine the most suitable number of permutation candidate sequences to achieve a reasonable PAPR reduction performance without increasing the time and hardware complexity to levels that the systems cannot tolerate.
Reducing water absorption characteristics of kraft paper reinforced with modified nanoparticles
(University of the Witwatersrand, Johannesburg, 2023) Katun, Mohammed Mahmood; Nyamupangedengu, Cuthbert; Gomes, Chandima
The inherent hydrophilic characteristics of cellulose wood fibre compromise the dielectric properties of kraft paper insulation that is used mainly in oil-insulated power transformers. This thesis, therefore, presents a novel material design model of nanocomposite kraft paper with improved hydrophobic properties for power transformer insulation applications. The concept of nanodielectric kraft paper design was used. Rutile-titanium dioxide nanoparticles (rutile-TiO2 NPs) were selected as the nanofiller. Compared with other metal-oxides, rutileTiO2 NPs are stable in chemical reactions, have good thermal stability and also have high electrical resistivity. Rutile-TiO2 of 19.72 nm diameter were fabricated using the sol-gel method and then used in reinforcing the kraft paper to produce a nanocomposite kraft paper with improved dielectric properties. Since the nanoparticles are inherently hydrophilic, and the intention is to produce a hydrophilic nanocomposite kraft paper, a technique was devised to make the NPs hydrophobic. The rutile-TiO2 NPs were surface conditioned with two alternative surfactants; Alkyl ketene dimer (AKD) and alkenyl succinic anhydride (ASA). Various quantities of the two surfactants were investigated to determine the optimal amount. The resultant surface-modified rutile-TiO2 NPs were studied to understand their hydrophilicity and thermal stability properties. It was found that the unmodified rutile-TiO2 NPs absorbed more moisture, compared with the surface-modified nanofiller, the mass increased by about 4% due to moisture absorption. The surface-modified rutile-TiO2 NPs with 5% AKD had 45% more thermal resilience than the unmodified rutile-TiO2 NPs surface and this is a significant knowledge contribution of this thesis. Using an unbleached kraft pulp, nanocomposite kraft paper specimens were fabricated with nanoparticles having varying surfactant loading. The specimens were then characterized to reveal various critical properties such as hydrophobicity, thermal, dielectric losses and dielectric strength. The version of the nanocomposite kraft paper that gave the most improved water and moisture absorption properties was that with 5vol/vol% ASA surface-modified NPs. The moisture absorption rate dropped by 74% compared to the unfilled kraft paper, and the water vapor transmission rate decreased by 30%. The contact angle of water droplets improved by 12%, and water absorption rate improved by being 4 times slower. The dielectric loss measurements showed that the nanocomposite kraft paper containing rutile-TiO2 NPs (5 vol/vol% ASA) also had 40% lower dielectric losses than the reference (unfilled) samples. The breakdown voltage of the nanocomposite kraft paper increased by about 15% while thermal withstand was improved by 5.4%. This research, therefore, has successfully improved the hydrophobic properties of kraft paper by filling it with surface-modified rutileTiO2 NP. It can be argued from the results that for power transformer application, the novel nanocomposite kraft paper developed in this thesis will improve power transformer insulation reliability design by mitigating the main agents of insulation degradation which are water and thermal stress.