7103 research outputs found
Sort by
Machine learning enhanced radio location of partial discharge
Get PDFPartial Discharge (PD) is a well-known indicator of plant failure in electricity facilities. A considerable proportion of assets including transformers, switch gears, and power lines are susceptible to PD due to incipient weakness of their dielectric components. These discharges may cause further degradation of the insulation, which in turn may lead to subsequent catastrophic failure. The damage that results from PD activity is worth millions of pounds and endangers the lives of personnel. PD emits electrical pulses in the form of Radio Frequency (RF) signals which propagate as a travelling wave in the vicinity of the discharge site and can be detected using dedicated sensors. This has motivated the use of an enhanced radio-based technique to detect its occurrence at early stage. Early detection of PD helps utility operators to initiate an emergency maintenance outside the scheduled times when it is most cost-effective and before the equipment loses performance or suffers catastrophic failure, hence improving asset management. Therefore this thesis presents an investigation of an enhanced machine learning approach to continuous PD localisation using a network of radio sensors. The approach being investigated relies on location dependent parameters which will be extracted from PD measurements. This thesis demonstrates RF-based fingerprinting technique for locating PD sources using Received Signal Strength (RSS). Furthermore, Signal Strength Ratios (SSR) between pairs of sensor nodes are used as robust fingerprints given that the energy emitted by each PD event may be different due to progressive nature of PD severity as deterioration continues and the fact that different types of PD occur in nature. Sophisticated machine learning techniques are investigated and used to develop PD localisation models. This work also investigates the plausibility of using other PD received signal parameters for locating PD sources. It has been found that the statistical characterisation of the received RF signals produces manifold PD features beside RSS. The developed localisation approach based on the analysis of these statistical features assumes that PDs generate unique RF spatial patterns due to the complexities and non-linearities of RF propagation. This approach exploits two distinct frequency bands which hold different PD information. PD location features are extracted from the main PD signal and the two sub-band signals. These features are then used to infer PD location. Moreover, due to the increased dimensionality of data that may result from PD feature generation, feature selection algorithm; Correlation Based Feature Selection (CFS) is employed for feature selection and dimensionality reduction. The use of statistical PD features improves localisation accuracy. This study further presents a novel method for RF-based PD localisation. The technique uses Wavelet Packet Transform (WPT) and machine learning ensemble methods to locate PDs. More specifically, the received signals are decomposed by the Maximal Overlap Discrete Wavelet Packet Transform (MODWPT) version of wavelet packet and analysed in order to identify localised PD signal patterns. The Regression Tree algorithm, Bootstrap Aggregating method and Regression Random Forest (RRF) are used to develop PD localisation models based on the wavelet PD features. The proposed PD localisation scheme has been found to successfully locate PD with negligible error. Additionally, the principle of the developed PD localisation system has been validated using a separate test dataset. This approach is based on purely practical reasons, given the enormity of separate experiments to be carried out. The data required is collated over an extended time period. The results of the investigation presented in this thesis show that an autonomous andefficient substation-wide RF-based continuous PD localisation system is possible.Partial Discharge (PD) is a well-known indicator of plant failure in electricity facilities. A considerable proportion of assets including transformers, switch gears, and power lines are susceptible to PD due to incipient weakness of their dielectric components. These discharges may cause further degradation of the insulation, which in turn may lead to subsequent catastrophic failure. The damage that results from PD activity is worth millions of pounds and endangers the lives of personnel. PD emits electrical pulses in the form of Radio Frequency (RF) signals which propagate as a travelling wave in the vicinity of the discharge site and can be detected using dedicated sensors. This has motivated the use of an enhanced radio-based technique to detect its occurrence at early stage. Early detection of PD helps utility operators to initiate an emergency maintenance outside the scheduled times when it is most cost-effective and before the equipment loses performance or suffers catastrophic failure, hence improving asset management. Therefore this thesis presents an investigation of an enhanced machine learning approach to continuous PD localisation using a network of radio sensors. The approach being investigated relies on location dependent parameters which will be extracted from PD measurements. This thesis demonstrates RF-based fingerprinting technique for locating PD sources using Received Signal Strength (RSS). Furthermore, Signal Strength Ratios (SSR) between pairs of sensor nodes are used as robust fingerprints given that the energy emitted by each PD event may be different due to progressive nature of PD severity as deterioration continues and the fact that different types of PD occur in nature. Sophisticated machine learning techniques are investigated and used to develop PD localisation models. This work also investigates the plausibility of using other PD received signal parameters for locating PD sources. It has been found that the statistical characterisation of the received RF signals produces manifold PD features beside RSS. The developed localisation approach based on the analysis of these statistical features assumes that PDs generate unique RF spatial patterns due to the complexities and non-linearities of RF propagation. This approach exploits two distinct frequency bands which hold different PD information. PD location features are extracted from the main PD signal and the two sub-band signals. These features are then used to infer PD location. Moreover, due to the increased dimensionality of data that may result from PD feature generation, feature selection algorithm; Correlation Based Feature Selection (CFS) is employed for feature selection and dimensionality reduction. The use of statistical PD features improves localisation accuracy. This study further presents a novel method for RF-based PD localisation. The technique uses Wavelet Packet Transform (WPT) and machine learning ensemble methods to locate PDs. More specifically, the received signals are decomposed by the Maximal Overlap Discrete Wavelet Packet Transform (MODWPT) version of wavelet packet and analysed in order to identify localised PD signal patterns. The Regression Tree algorithm, Bootstrap Aggregating method and Regression Random Forest (RRF) are used to develop PD localisation models based on the wavelet PD features. The proposed PD localisation scheme has been found to successfully locate PD with negligible error. Additionally, the principle of the developed PD localisation system has been validated using a separate test dataset. This approach is based on purely practical reasons, given the enormity of separate experiments to be carried out. The data required is collated over an extended time period. The results of the investigation presented in this thesis show that an autonomous andefficient substation-wide RF-based continuous PD localisation system is possible
Executives' characteristics and corporate policies
Get PDFRecent regulatory and demographic developments in corporate governance point to changes in managerial gender and age. These changes call for a better understanding of the various implications of managers' gender and age on firms. Existing studies suggest that managerial gender and age are systematically related to the riskiness of the firm (Huang and Kisgen 2013; Serfling 2014). These studies are mainly focused on the CEO. However, Hambrick and Mason (1984) advance the Upper Echelons Theory (UET) and argue that management is a shared activity, in which CEOs delegate responsibilities and authority to the rest of their top management teams (TMTs). They call for examining the implications of managerial characteristics at the TMT level, rather than the CEO alone. This thesis extends the literature by analysing the predictions of the UET, specifically by empirically examining whether the proportion of female managers in the TMT and the average age of the TMT are related to three major corporate financial policies.The first empirical chapter (Chapter 3) investigates corporate cash holdings, which are held for precautionary reasons (Bates, Kahle, and Stulz 2009). Hence, I examine whether managerial gender and age at the TMT level are related to cash holdings. Using a large sample comprising S&P 1500 firms for the period 1992 to 2013, the results indicate that the percentage of female managers in the TMT is positively related to cash holdings, indicating a possibility that TMTs with more female executives are more risk-averse or less overconfident, leading to further emphasis on the precautionary need for cash. Further, the average age of the TMT is negatively related to cash holdings, suggesting that older TMTs are more confident with regard to taking risker choices. Further analysis suggests that the ages of the CEO and CFO are negatively related to cash holdings, indicating that executives other than the CEO may exercise some influence over the cash holdings decision.;The second empirical chapter (Chapter 4) studies R&D investments, which are risky since they entail certain costs with highly uncertain payoffs (Abdel-Khalik 2014). Therefore, I investigate whether managers' gender and age at the TMT level are related to R&D investments. The analyses indicate that the percentage of female managers in the TMT and R&D investments are negatively related, consistent with the view that TMTs with more females might be more risk-averse or less overconfident, reducing investments in risky assets. Also, both female CEOs and CFOs are negatively related to R&D investments. Moreover, the results point to the lack of a systematic relationship between the average age of the TMT and R&D investments, which is theoretically possible since the experience we gain from ageing could offset the risk-aversion we develop as we age (Worthy et al. 2011). Nevertheless, the additional analysis shows that the CEO's age (CFO's age) is negatively (positively) related to R&D investments. Thus, the lack of a systematic relation between the average age of the TMT and R&D investments might be a product of cancelling out effect within the members of the TMT.;The third empirical chapter (Chapter 5) examines two aspects of corporate payout policy; namely, the payout levels and payout methods. First, managers may face trade-offs between payouts and investments, assuming that they cannot accumulate cash for perpetuity (Caliskan and Doukas 2015). Thus, I examine whether managerial gender and age at the TMT level are related to the corporate payout levels. The analysis indicates that the percentage of female managers in the TMT is positively related to the payout levels. This is possibly because TMTs with more females are more risk-averse or less overconfident, choosing to distribute funds, given that that this choice is less risky than investments. There is also some evidence to suggest that the average age of the TMT and payout levels are negatively related, indicating that older TMTs take more risks (i.e. invest rather than distribute funds). Further analysis reveals some evidence that the CFO's age is negatively associated with the payout levels, suggesting that CFOs are more influential in setting the payout levels.Second, after setting the payout amount, managers decide on the payout method; broadly, dividends or stock repurchases (Eisdorfer, Giaccotto, and White 2015). Compared to dividends, stock repurchases improve the financial flexibility of the firm since they do not entail future commitments (Bonaimé, Hankins, and Harford 2014). Accordingly, I investigate whether managerial gender and age at the TMT level are related to the corporate payout method. The analysis shows that the proportion of female executives in the TMT and the proportion of stock repurchase to total payouts are positively related. This is possible because such TMTs rely more on stock repurchases in distributing cash to the shareholders to retain the financial flexibility of the firm. The average age of the TMT is negatively related to the payout flexibility, indicating that older TMTs may adopt risker choices. Additional analysis shows that the ages of the CEOs and CFOs are negatively related to payout flexibility, supporting the view that managers other than the CEO could influence the corporate policies and highlighting the importance of incorporating other TMT members when examining managerial characteristics.Recent regulatory and demographic developments in corporate governance point to changes in managerial gender and age. These changes call for a better understanding of the various implications of managers' gender and age on firms. Existing studies suggest that managerial gender and age are systematically related to the riskiness of the firm (Huang and Kisgen 2013; Serfling 2014). These studies are mainly focused on the CEO. However, Hambrick and Mason (1984) advance the Upper Echelons Theory (UET) and argue that management is a shared activity, in which CEOs delegate responsibilities and authority to the rest of their top management teams (TMTs). They call for examining the implications of managerial characteristics at the TMT level, rather than the CEO alone. This thesis extends the literature by analysing the predictions of the UET, specifically by empirically examining whether the proportion of female managers in the TMT and the average age of the TMT are related to three major corporate financial policies.The first empirical chapter (Chapter 3) investigates corporate cash holdings, which are held for precautionary reasons (Bates, Kahle, and Stulz 2009). Hence, I examine whether managerial gender and age at the TMT level are related to cash holdings. Using a large sample comprising S&P 1500 firms for the period 1992 to 2013, the results indicate that the percentage of female managers in the TMT is positively related to cash holdings, indicating a possibility that TMTs with more female executives are more risk-averse or less overconfident, leading to further emphasis on the precautionary need for cash. Further, the average age of the TMT is negatively related to cash holdings, suggesting that older TMTs are more confident with regard to taking risker choices. Further analysis suggests that the ages of the CEO and CFO are negatively related to cash holdings, indicating that executives other than the CEO may exercise some influence over the cash holdings decision.;The second empirical chapter (Chapter 4) studies R&D investments, which are risky since they entail certain costs with highly uncertain payoffs (Abdel-Khalik 2014). Therefore, I investigate whether managers' gender and age at the TMT level are related to R&D investments. The analyses indicate that the percentage of female managers in the TMT and R&D investments are negatively related, consistent with the view that TMTs with more females might be more risk-averse or less overconfident, reducing investments in risky assets. Also, both female CEOs and CFOs are negatively related to R&D investments. Moreover, the results point to the lack of a systematic relationship between the average age of the TMT and R&D investments, which is theoretically possible since the experience we gain from ageing could offset the risk-aversion we develop as we age (Worthy et al. 2011). Nevertheless, the additional analysis shows that the CEO's age (CFO's age) is negatively (positively) related to R&D investments. Thus, the lack of a systematic relation between the average age of the TMT and R&D investments might be a product of cancelling out effect within the members of the TMT.;The third empirical chapter (Chapter 5) examines two aspects of corporate payout policy; namely, the payout levels and payout methods. First, managers may face trade-offs between payouts and investments, assuming that they cannot accumulate cash for perpetuity (Caliskan and Doukas 2015). Thus, I examine whether managerial gender and age at the TMT level are related to the corporate payout levels. The analysis indicates that the percentage of female managers in the TMT is positively related to the payout levels. This is possibly because TMTs with more females are more risk-averse or less overconfident, choosing to distribute funds, given that that this choice is less risky than investments. There is also some evidence to suggest that the average age of the TMT and payout levels are negatively related, indicating that older TMTs take more risks (i.e. invest rather than distribute funds). Further analysis reveals some evidence that the CFO's age is negatively associated with the payout levels, suggesting that CFOs are more influential in setting the payout levels.Second, after setting the payout amount, managers decide on the payout method; broadly, dividends or stock repurchases (Eisdorfer, Giaccotto, and White 2015). Compared to dividends, stock repurchases improve the financial flexibility of the firm since they do not entail future commitments (Bonaimé, Hankins, and Harford 2014). Accordingly, I investigate whether managerial gender and age at the TMT level are related to the corporate payout method. The analysis shows that the proportion of female executives in the TMT and the proportion of stock repurchase to total payouts are positively related. This is possible because such TMTs rely more on stock repurchases in distributing cash to the shareholders to retain the financial flexibility of the firm. The average age of the TMT is negatively related to the payout flexibility, indicating that older TMTs may adopt risker choices. Additional analysis shows that the ages of the CEOs and CFOs are negatively related to payout flexibility, supporting the view that managers other than the CEO could influence the corporate policies and highlighting the importance of incorporating other TMT members when examining managerial characteristics
Wind turbine gearbox diagnostics using artificial intelligence
No full textTo meet the latest strike price, the cost of energy from wind turbines needs to decrease.One of the biggest cost contributors to wind energy is the operation and maintenance cost. If this cost is driven down, the cost of energy of wind will substantially decrease and the reliability of the wind turbine assets need to increase. For that reason, condition monitoring systems are installed in modern wind turbines. These systems collect data and in abnormal conditions trigger alarms that are an indication of a fault. Maintenance actions can be scheduled accordingly that way, and faulty components can be replaced before catastrophic failures and large downtimes occur.Therefore, the aim of this thesis is to utilise vibration and performance data collected from wind turbine gearboxes, in order to perform fault detection and diagnosis.The data is collected at various times prior to gearbox component failures and advanced signal processing techniques are applied to reveal fault signatures. Machine learning models are trained based on features extracted from vibration spectra and operational data separately, but also a combination of these two types of data is investigated. The output is fault detection and isolation on gearbox component level. Both unit-specific and fleet-based methods are examines. The models are trained on specific turbines,but the generalization to other turbines is also examined.The above will provide a exible but robust framework for the early detection of emerging wind turbine faults. This will lead to minimisation of the wind turbine downtime and increase of the wind turbines reliability and income through operational enhancement.To meet the latest strike price, the cost of energy from wind turbines needs to decrease.One of the biggest cost contributors to wind energy is the operation and maintenance cost. If this cost is driven down, the cost of energy of wind will substantially decrease and the reliability of the wind turbine assets need to increase. For that reason, condition monitoring systems are installed in modern wind turbines. These systems collect data and in abnormal conditions trigger alarms that are an indication of a fault. Maintenance actions can be scheduled accordingly that way, and faulty components can be replaced before catastrophic failures and large downtimes occur.Therefore, the aim of this thesis is to utilise vibration and performance data collected from wind turbine gearboxes, in order to perform fault detection and diagnosis.The data is collected at various times prior to gearbox component failures and advanced signal processing techniques are applied to reveal fault signatures. Machine learning models are trained based on features extracted from vibration spectra and operational data separately, but also a combination of these two types of data is investigated. The output is fault detection and isolation on gearbox component level. Both unit-specific and fleet-based methods are examines. The models are trained on specific turbines,but the generalization to other turbines is also examined.The above will provide a exible but robust framework for the early detection of emerging wind turbine faults. This will lead to minimisation of the wind turbine downtime and increase of the wind turbines reliability and income through operational enhancement
Empirical investigations of foreign equity investments in an emerging market
Get PDFThis thesis examines foreign institutional (portfolio) investors' (FIIs or FIIs) influence on government policy, their trading behaviour, and their implications on firm-level board monitoring in a large emerging market.In the first empirical investigation, we examine the power of FIIs to directly influence the policy of host government. In a quasi-experimental setting, we find a strong negative stock market reaction to an exogenous policy shock that threatens to increase tax liability of FIIs in India. More importantly, the shock resulted in a daily market withdrawal of approximately 0.309 basis points of market capitalization for an average equity by FIIs. Further, the results also indicate that FIIs' withdrawal has a disruptive effect on several aspects of the stock market including volatility, liquidity and prices. Finally, the effect of the shock seems to have a long-term detrimental effect as, after the tax threat is removed, FIIs do not re-enter the market with the same speed and volume of trading compared to the initial market withdrawal.In the second empirical investigation, we examine the information content of opportunistic and routine insiders' trades in an emerging market and test whether foreign institutional investors (FIIs) exploit and mimic informative trades. We find that opportunistic trades translate into an incremental return of approximately 243 basis points in the following month of the trade, much higher than previously reported in developed markets. More importantly, by exploiting unique high-frequency trade-level transaction data, we find that FIIs mimic past opportunistic insiders' buy trades and earn superior abnormal returns. In sum, this study implies that opportunistic insiders' trading in emerging markets enables FIIs to reduce their informational disadvantage.;In the third empirical investigation, we explore whether FIIs improve board monitoring. Exploiting the global financial crisis of 2007-08 as an exogenous shock that resulted in a significant decline of FIIs' ownership in the Indian market, we find evidence of a causal link between FIIs' ownership and different dimensions of board monitoring. Specifically, the empirical results suggest that FIIs reduce board size,busyness, network size, CEO power, and CEO pay, and improve board diligence. However, we also document a negative link between FIIs' ownership and board independence, indicating FIIs may not view independent directors as effective monitors in this market.This thesis examines foreign institutional (portfolio) investors' (FIIs or FIIs) influence on government policy, their trading behaviour, and their implications on firm-level board monitoring in a large emerging market.In the first empirical investigation, we examine the power of FIIs to directly influence the policy of host government. In a quasi-experimental setting, we find a strong negative stock market reaction to an exogenous policy shock that threatens to increase tax liability of FIIs in India. More importantly, the shock resulted in a daily market withdrawal of approximately 0.309 basis points of market capitalization for an average equity by FIIs. Further, the results also indicate that FIIs' withdrawal has a disruptive effect on several aspects of the stock market including volatility, liquidity and prices. Finally, the effect of the shock seems to have a long-term detrimental effect as, after the tax threat is removed, FIIs do not re-enter the market with the same speed and volume of trading compared to the initial market withdrawal.In the second empirical investigation, we examine the information content of opportunistic and routine insiders' trades in an emerging market and test whether foreign institutional investors (FIIs) exploit and mimic informative trades. We find that opportunistic trades translate into an incremental return of approximately 243 basis points in the following month of the trade, much higher than previously reported in developed markets. More importantly, by exploiting unique high-frequency trade-level transaction data, we find that FIIs mimic past opportunistic insiders' buy trades and earn superior abnormal returns. In sum, this study implies that opportunistic insiders' trading in emerging markets enables FIIs to reduce their informational disadvantage.;In the third empirical investigation, we explore whether FIIs improve board monitoring. Exploiting the global financial crisis of 2007-08 as an exogenous shock that resulted in a significant decline of FIIs' ownership in the Indian market, we find evidence of a causal link between FIIs' ownership and different dimensions of board monitoring. Specifically, the empirical results suggest that FIIs reduce board size,busyness, network size, CEO power, and CEO pay, and improve board diligence. However, we also document a negative link between FIIs' ownership and board independence, indicating FIIs may not view independent directors as effective monitors in this market
Disability hate crime and social work
Get PDFAccording to statistics from the Crown Office and Procurator Fiscal Service, instances of disability hate crime in Scotland have been rising since 2010. The reasons for this have yet to be fully explored in research, although there is a strong belief amongst those working with disabled people, that the vast majority of such incidents go unreported. This study explores disability hate crime conceptually, and in practice, through the testimony of disabled people themselves, and social workers who work alongside them. Utilising Social Relational Model of disability, alongside a methodology influenced by Interpretative Phenomenological Analysis, the study aims to fill the gap in research by positioning disabled people's voices at the forefront of the data collected and analysed. By conducting interviews with disabled people, social workers, and disabled people's organisations in central Scotland, the study finds that an individual's relationship with a disability identity can play a significant impact in how disabled engage with disability hate crime as a concept, and that while social workers remain enthusiastic and supportive of disabled people who are experiencing disability hate crimes, they themselves suggest that they are being hindered in their ability to offer their best practice, as pressures of time, money, and management, are inhibiting their efforts in the area.According to statistics from the Crown Office and Procurator Fiscal Service, instances of disability hate crime in Scotland have been rising since 2010. The reasons for this have yet to be fully explored in research, although there is a strong belief amongst those working with disabled people, that the vast majority of such incidents go unreported. This study explores disability hate crime conceptually, and in practice, through the testimony of disabled people themselves, and social workers who work alongside them. Utilising Social Relational Model of disability, alongside a methodology influenced by Interpretative Phenomenological Analysis, the study aims to fill the gap in research by positioning disabled people's voices at the forefront of the data collected and analysed. By conducting interviews with disabled people, social workers, and disabled people's organisations in central Scotland, the study finds that an individual's relationship with a disability identity can play a significant impact in how disabled engage with disability hate crime as a concept, and that while social workers remain enthusiastic and supportive of disabled people who are experiencing disability hate crimes, they themselves suggest that they are being hindered in their ability to offer their best practice, as pressures of time, money, and management, are inhibiting their efforts in the area
Acoustic-based assistive technology tools for dysarthria managment
Get PDFThe research presented in this thesis addresses the concepts of application of important digital signal processing algorithms in the detection and treatment of dysarthria, a neurological motor speech disorder. The novel algorithms presented in this thesis include a silence, unvoiced and voiced segmentation technique for dysarthric speech based on linear prediction error variance (LPEV), an automatic diadochokinetic (DDK) analysis and segmentation scheme for dysarthric speech, the application of speech processing algorithms in the extraction of prosodic, voice quality, pronunciation and wavelet features for the detection and severity classification in dysarthric speech and the modification of dysarthric speech features using speech enhancement techniques to improve the intelligibility of dysarthric speech in a stress production exercise for the treatment of dysarthria.In particular, an improved silence, unvoiced and voiced segmentation technique for dysarthric speech is proposed. This method is an enhanced technique that makes use of a two-layer segmentation approach which combines the short-time-energy (STE) and LPEV to distinctly differentiate between the silence and voiced segments despite the reduced/inconsistent intensity, pauses, voice breaks and slow speech rate experienced in dysarthric speech. Including the LPEV into the segmentation process has proved to be advantageous in eliminating segmentation errors due to the similarity observed between the STE profiles of the silence and voiced segments in dysarthric speech. The experimental results have shown that this segmentation method is also effective and efficient in reducing the effects of artefacts introduced in dysarthric speech.;A novel automatic DDK analysis scheme is proposed in this research to extract individual DDK syllables and analyse them for consistency. This method is based on a speaker-specific moving average threshold (rather than a fixed threshold) which addresses the varying intensities in the DDK sounds produced by speakers with dysarthria. This method also addresses the challenge of intra-syllable breaks introduced in dysarthric DDK syllables using a minimum distance merging approach. In addition, the algorithm analyses the segmented DDK syllables by calculating the individual DDK rates and their variance in order to measure the DDK syllable production consistency. The high accuracy of the proposed method is tested and verified using both dysarthric and healthy controlled databases.Three novel schemes for automatic detection and severity classification of the dysarthric speech are also proposed in this research. One extracts an extended speech feature called centroid formant (which is a representation of energy concentration in the frequency spectrum) and classifies these centroid formants using neural network classifiers for the detection of dysarthria. The centroid formant-based detection scheme also forms the backbone for the development of the second and more robust detection scheme which combines centroid formants with prosodic, voice quality, pronunciation and wavelet features for more efficient classification. A third scheme is developed specifically for the classification of dysarthria into three severity levels using the same features as in the second scheme. The efficiencies of these detection and severity classification schemes are evaluated by calculating the accuracy, sensitivity and specificity of the classifiers.The effects of modification of prosodic cues used in stress production on the ability of listeners to correctly identify the position of the stressed word in sentences are also investigated in this research. This investigation is focused on the three prosodic cues used by healthy controlled speakers in stress production; namely intensity, duration and fundamental frequency. These three features are modified acoustically and presented to untrained listeners in an aim to evaluate the effects of the individual and combined modifications on the listeners' perception. The findings of this investigation will help clinicians, including speech and language therapists, make an informed decision on the prosodic feature to focus on during stress production exercises for the management of dysarthria.Finally, the dysarthria management schemes proposed in this research are developed into user-interactive tools in MATLAB from which speaker-specific information and reports can be generated and downloaded for progress monitoring and further analytical purposes.The research presented in this thesis addresses the concepts of application of important digital signal processing algorithms in the detection and treatment of dysarthria, a neurological motor speech disorder. The novel algorithms presented in this thesis include a silence, unvoiced and voiced segmentation technique for dysarthric speech based on linear prediction error variance (LPEV), an automatic diadochokinetic (DDK) analysis and segmentation scheme for dysarthric speech, the application of speech processing algorithms in the extraction of prosodic, voice quality, pronunciation and wavelet features for the detection and severity classification in dysarthric speech and the modification of dysarthric speech features using speech enhancement techniques to improve the intelligibility of dysarthric speech in a stress production exercise for the treatment of dysarthria.In particular, an improved silence, unvoiced and voiced segmentation technique for dysarthric speech is proposed. This method is an enhanced technique that makes use of a two-layer segmentation approach which combines the short-time-energy (STE) and LPEV to distinctly differentiate between the silence and voiced segments despite the reduced/inconsistent intensity, pauses, voice breaks and slow speech rate experienced in dysarthric speech. Including the LPEV into the segmentation process has proved to be advantageous in eliminating segmentation errors due to the similarity observed between the STE profiles of the silence and voiced segments in dysarthric speech. The experimental results have shown that this segmentation method is also effective and efficient in reducing the effects of artefacts introduced in dysarthric speech.;A novel automatic DDK analysis scheme is proposed in this research to extract individual DDK syllables and analyse them for consistency. This method is based on a speaker-specific moving average threshold (rather than a fixed threshold) which addresses the varying intensities in the DDK sounds produced by speakers with dysarthria. This method also addresses the challenge of intra-syllable breaks introduced in dysarthric DDK syllables using a minimum distance merging approach. In addition, the algorithm analyses the segmented DDK syllables by calculating the individual DDK rates and their variance in order to measure the DDK syllable production consistency. The high accuracy of the proposed method is tested and verified using both dysarthric and healthy controlled databases.Three novel schemes for automatic detection and severity classification of the dysarthric speech are also proposed in this research. One extracts an extended speech feature called centroid formant (which is a representation of energy concentration in the frequency spectrum) and classifies these centroid formants using neural network classifiers for the detection of dysarthria. The centroid formant-based detection scheme also forms the backbone for the development of the second and more robust detection scheme which combines centroid formants with prosodic, voice quality, pronunciation and wavelet features for more efficient classification. A third scheme is developed specifically for the classification of dysarthria into three severity levels using the same features as in the second scheme. The efficiencies of these detection and severity classification schemes are evaluated by calculating the accuracy, sensitivity and specificity of the classifiers.The effects of modification of prosodic cues used in stress production on the ability of listeners to correctly identify the position of the stressed word in sentences are also investigated in this research. This investigation is focused on the three prosodic cues used by healthy controlled speakers in stress production; namely intensity, duration and fundamental frequency. These three features are modified acoustically and presented to untrained listeners in an aim to evaluate the effects of the individual and combined modifications on the listeners' perception. The findings of this investigation will help clinicians, including speech and language therapists, make an informed decision on the prosodic feature to focus on during stress production exercises for the management of dysarthria.Finally, the dysarthria management schemes proposed in this research are developed into user-interactive tools in MATLAB from which speaker-specific information and reports can be generated and downloaded for progress monitoring and further analytical purposes
The role of Minecraft on social-emotional and behavioural outcomes of children with hearing loss or Autism : perspectives of parents and children
Get PDFChildren with Autism Spectrum Disorder (ASD) or Hearing Loss (HL) have relationship challenges and mental health difficulties due to functional disturbance affecting social interaction. This study examined the role of Online Computer Game (OCG), and specifically Minecraft (MC), to facilitate social relations, mental health, and the wellbeing of children with ASD and/or HL in the United Kingdom (UK) and the Kingdom of Saudi Arabia (KSA). MC is a sandbox computer game in open-world format and recognised to be socially interactive gameplay, chosen here due to its popularity, accessibility and cooperative gameplay characteristics. In the first phase of this research, a systematic literature review was conducted of all peer-reviewed articles that were written in English and included first-hand evidence to synthesise the evidence for and against MC use in education (n=38). The review concluded MC to be beneficial to children regarding increased motivation for academic learning and social development including communication, sharing and collaboration skills. Therefore, the second phase was conducted to identify correlations between playing OCG or MC and children's social-emotional and behavioural outcomes, and specifically players' peer relationship problems using the convergent mixed methods design approach. Data consisted of three parts: questionnaire (n=255), interviews (n=7) and observations (n=4). Subjects for the questionnaire were parents of primary school children aged 8 and over from three groups: children with ASD (n=121), children with HL (n=11) and Typical Developing (TD) children (n=123). This thesis reported that MC is a social or entertaining activity that can be used as a place for social intervention for three reasons. First, cooperative gameplay on MC has no significant associations with difficulties on the SDQ for either TD or children with ASD in this research sample. Secondly, higher frequency of playing MC with others is associated with a lower peer relationship problems score in the KSA sample. Thirdly, the qualitative pieces of evidence show that the benefits outweigh the risks of playing MC, notably for children with ASD or HL. Therefore, MC might be potentially beneficial for social intervention for children with ASD or HL. Parents reported three main reasons for being interested in MC for children with ASD or HL: peer relationships and peer support (i.e., a space for social interaction with others), emotional benefits (e.g., enjoyment and being happy) and behaviour benefits (i.e.,being calm and relaxed or as a reward for desirable behaviours). Concerns about addiction, safety, and physical activity use were raised, but evidence shows that most of these concerns are related to gaming management rather than MC itself as a game. Concerns and thesis' limitations are discussed. Altogether, these data suggest that MC game-play may be considered appropriate for social interventions for children with ASD or HL, and may be considered for incorporation into educational pedagogy or psychological support for its social benefits. The study significantly added understanding of gaming and diagnostic condition characteristics in the role of peer relationship skills among children. The findings may help to advance current literature in the areas of children's social-emotional and behavioural development.Children with Autism Spectrum Disorder (ASD) or Hearing Loss (HL) have relationship challenges and mental health difficulties due to functional disturbance affecting social interaction. This study examined the role of Online Computer Game (OCG), and specifically Minecraft (MC), to facilitate social relations, mental health, and the wellbeing of children with ASD and/or HL in the United Kingdom (UK) and the Kingdom of Saudi Arabia (KSA). MC is a sandbox computer game in open-world format and recognised to be socially interactive gameplay, chosen here due to its popularity, accessibility and cooperative gameplay characteristics. In the first phase of this research, a systematic literature review was conducted of all peer-reviewed articles that were written in English and included first-hand evidence to synthesise the evidence for and against MC use in education (n=38). The review concluded MC to be beneficial to children regarding increased motivation for academic learning and social development including communication, sharing and collaboration skills. Therefore, the second phase was conducted to identify correlations between playing OCG or MC and children's social-emotional and behavioural outcomes, and specifically players' peer relationship problems using the convergent mixed methods design approach. Data consisted of three parts: questionnaire (n=255), interviews (n=7) and observations (n=4). Subjects for the questionnaire were parents of primary school children aged 8 and over from three groups: children with ASD (n=121), children with HL (n=11) and Typical Developing (TD) children (n=123). This thesis reported that MC is a social or entertaining activity that can be used as a place for social intervention for three reasons. First, cooperative gameplay on MC has no significant associations with difficulties on the SDQ for either TD or children with ASD in this research sample. Secondly, higher frequency of playing MC with others is associated with a lower peer relationship problems score in the KSA sample. Thirdly, the qualitative pieces of evidence show that the benefits outweigh the risks of playing MC, notably for children with ASD or HL. Therefore, MC might be potentially beneficial for social intervention for children with ASD or HL. Parents reported three main reasons for being interested in MC for children with ASD or HL: peer relationships and peer support (i.e., a space for social interaction with others), emotional benefits (e.g., enjoyment and being happy) and behaviour benefits (i.e.,being calm and relaxed or as a reward for desirable behaviours). Concerns about addiction, safety, and physical activity use were raised, but evidence shows that most of these concerns are related to gaming management rather than MC itself as a game. Concerns and thesis' limitations are discussed. Altogether, these data suggest that MC game-play may be considered appropriate for social interventions for children with ASD or HL, and may be considered for incorporation into educational pedagogy or psychological support for its social benefits. The study significantly added understanding of gaming and diagnostic condition characteristics in the role of peer relationship skills among children. The findings may help to advance current literature in the areas of children's social-emotional and behavioural development
A novel meta-synthesis approach to technology maturity management for project risk control
Get PDFWith the rapid development of science and technology in modern society, "high-tech" is becoming the main characteristic of complex systems. However, due to the large scale of projects, and the intricate architecture and functionalities involved in these systems,traditional project development processes often encounter schedule delays, cost increases or performance degradation. Most of these are related to technical risk. Moreover, rapid change in both domestic and international markets and the acceleration of technology upgrading are bringing more challenges to risk management for complex system development.To improve the risk management of key technology developments aiming to ensure the success of technical-critical complex systems, a Meta-Synthesis Approach (MSA) with technology assessment is proposed in this thesis. It combines quantitative and qualitative methods for assessing the maturity of technology (Technology Readiness Level, TRL) andalso the difficulty of enhancing the technology maturity (Advancement Degree of Difficulty,AD2). The research focuses on the qualitative analysis of TRL assessment and quantitative assessment of AD2. Concepts and characteristics of technology maturity are discussed. An attributes-based qualitative TRL assessment method was developed and a unified maturity evidence chain is defined to capture and manage the maturity assessment information. A genetic algorithm based Numerical Integration method was developed for the quantitative assessment of AD2, embodying key calculation of parameters for maturity weight factors.After that, a visualisation method with technology maturity diagrams is proposed. All the above methods were integrated into MSA to assist decision-making of technology risk management. The advantages of the MSA are then discussed, detailing the spiral development process of Machine-factor (data modelling, simulation and analysis) and Man-Factor (Expert Decision-making System) to produce an iterative refinement of the technical risk management. The optimised methods and process are supported by an information and procedure management software tool. To evaluate the approach, the author employed nine case studies and questionnaires to illustrate the usefulness and efficacy ofnew methods and integrated processes. The thesis concludes with a discussion of theresearch including the limitations of the results and future research.;The key technological innovations of the research conducted in this thesis are listed below:(1) A multi maturity attributes based qualitative TRL assessment method was proposed,demonstrating four views of the technology risk/maturity aspect, including (a) representation of the formation of the critical technology element state, (b) integration of the critical technology elements towards the system, (c) the fidelity of the testing or demonstration environments, together with (d) key performance indicators.(2) A computational method for calculating AD2 is presented. Based on the unified assessment evidence chain, the method establishes a numerical integration calculation of AD2. During the numerical integration, a genetic algorithm is used to optimise the assessment features weight factors.(3) A visualisation modelling based decision-making and information management software was developed in the research. It supports technology risk identification and the management process in aiming to facilitate technical decision-making, which is essential for a complex technology system management.(4) A systematic approach based on MSA is proposed, which combines qualitative and quantitative assessment methods. Assessment information and a procedural management software platform acting as a "Machine" are concatenated to an experts group acting as a"Man", the whole composing a "Man-Machine System". To enhance the veracity of the TRLand AD2 assessment and apply the assessments to gain insight of technology risks, the "Man"dominates the synthesis procedure, and the "Machine" carries out the supporting quantitative calculations.The multi-attributes qualitative TRL assessment method could not only identify the best practices origin of technical risk management in western society, but also combine Chinese System Engineering with an oriental system-view philosophy. The quantitative method,including the numerical integration calculation of AD2 proposed in this thesis, is helpful in building mathematical expressions of the technical difficulty and could provide a basis for further system R&D management. The above two methods are based on the unified assessment evidence chain. They are a first attempt to reveal the nature of technology risk.The limitation of the research is that the proposed quantitative method is based on the distinct decomposition of technology efforts and all the decoupled (when needed) efforts could map to a cardinal range value as they constitute a sample set for the numerical integration. In addition, the knowledge database could be embodied in the future to support the "Man-Machine" spiral advancement for understanding and improving the technology maturity.With the rapid development of science and technology in modern society, "high-tech" is becoming the main characteristic of complex systems. However, due to the large scale of projects, and the intricate architecture and functionalities involved in these systems,traditional project development processes often encounter schedule delays, cost increases or performance degradation. Most of these are related to technical risk. Moreover, rapid change in both domestic and international markets and the acceleration of technology upgrading are bringing more challenges to risk management for complex system development.To improve the risk management of key technology developments aiming to ensure the success of technical-critical complex systems, a Meta-Synthesis Approach (MSA) with technology assessment is proposed in this thesis. It combines quantitative and qualitative methods for assessing the maturity of technology (Technology Readiness Level, TRL) andalso the difficulty of enhancing the technology maturity (Advancement Degree of Difficulty,AD2). The research focuses on the qualitative analysis of TRL assessment and quantitative assessment of AD2. Concepts and characteristics of technology maturity are discussed. An attributes-based qualitative TRL assessment method was developed and a unified maturity evidence chain is defined to capture and manage the maturity assessment information. A genetic algorithm based Numerical Integration method was developed for the quantitative assessment of AD2, embodying key calculation of parameters for maturity weight factors.After that, a visualisation method with technology maturity diagrams is proposed. All the above methods were integrated into MSA to assist decision-making of technology risk management. The advantages of the MSA are then discussed, detailing the spiral development process of Machine-factor (data modelling, simulation and analysis) and Man-Factor (Expert Decision-making System) to produce an iterative refinement of the technical risk management. The optimised methods and process are supported by an information and procedure management software tool. To evaluate the approach, the author employed nine case studies and questionnaires to illustrate the usefulness and efficacy ofnew methods and integrated processes. The thesis concludes with a discussion of theresearch including the limitations of the results and future research.;The key technological innovations of the research conducted in this thesis are listed below:(1) A multi maturity attributes based qualitative TRL assessment method was proposed,demonstrating four views of the technology risk/maturity aspect, including (a) representation of the formation of the critical technology element state, (b) integration of the critical technology elements towards the system, (c) the fidelity of the testing or demonstration environments, together with (d) key performance indicators.(2) A computational method for calculating AD2 is presented. Based on the unified assessment evidence chain, the method establishes a numerical integration calculation of AD2. During the numerical integration, a genetic algorithm is used to optimise the assessment features weight factors.(3) A visualisation modelling based decision-making and information management software was developed in the research. It supports technology risk identification and the management process in aiming to facilitate technical decision-making, which is essential for a complex technology system management.(4) A systematic approach based on MSA is proposed, which combines qualitative and quantitative assessment methods. Assessment information and a procedural management software platform acting as a "Machine" are concatenated to an experts group acting as a"Man", the whole composing a "Man-Machine System". To enhance the veracity of the TRLand AD2 assessment and apply the assessments to gain insight of technology risks, the "Man"dominates the synthesis procedure, and the "Machine" carries out the supporting quantitative calculations.The multi-attributes qualitative TRL assessment method could not only identify the best practices origin of technical risk management in western society, but also combine Chinese System Engineering with an oriental system-view philosophy. The quantitative method,including the numerical integration calculation of AD2 proposed in this thesis, is helpful in building mathematical expressions of the technical difficulty and could provide a basis for further system R&D management. The above two methods are based on the unified assessment evidence chain. They are a first attempt to reveal the nature of technology risk.The limitation of the research is that the proposed quantitative method is based on the distinct decomposition of technology efforts and all the decoupled (when needed) efforts could map to a cardinal range value as they constitute a sample set for the numerical integration. In addition, the knowledge database could be embodied in the future to support the "Man-Machine" spiral advancement for understanding and improving the technology maturity
Wind power utilization improvement by using hierarchical controlled microgrid
Get PDFIn this thesis, a possible solution for improving the utilization of wind power energy without causing impact to the power grid by using microgrid concept, which promises the function of making a cluster of generators and loads behave like a controllable unit, is proposed. This microgrid concept is achieved by a proposed hierarchical control method which contains three levels: decentralized level 1(primary or bottom level)is of fast response and can satisfy the essential power system operation requirements; centralized level 2 (secondary or middle level) eliminates control target deviation and rely on low band width communication channels to achieve accurate control objectives within several seconds; level 3(tertiary or top level) is the system organization level in which on-site functions such as seamless states transition operation and off-site functions such as setting points optimization are executed. A 4-bus microgrid model built in MATLAB/Simulink is used for individual validation tests of the different control level. The following experiments are executed in the wind power generators integrated CERTS 20-bus microgrid system. The simulation results show the system variations caused by wind power can be digested within the microgrid system with cooperation of the local loads, backup generators and energy storage devices which are operated by the hierarchical control system. In which case, the impact of wind power fluctuation can be absorbed by using microgrid concept with the proposed hierarchical control method.In this thesis, a possible solution for improving the utilization of wind power energy without causing impact to the power grid by using microgrid concept, which promises the function of making a cluster of generators and loads behave like a controllable unit, is proposed. This microgrid concept is achieved by a proposed hierarchical control method which contains three levels: decentralized level 1(primary or bottom level)is of fast response and can satisfy the essential power system operation requirements; centralized level 2 (secondary or middle level) eliminates control target deviation and rely on low band width communication channels to achieve accurate control objectives within several seconds; level 3(tertiary or top level) is the system organization level in which on-site functions such as seamless states transition operation and off-site functions such as setting points optimization are executed. A 4-bus microgrid model built in MATLAB/Simulink is used for individual validation tests of the different control level. The following experiments are executed in the wind power generators integrated CERTS 20-bus microgrid system. The simulation results show the system variations caused by wind power can be digested within the microgrid system with cooperation of the local loads, backup generators and energy storage devices which are operated by the hierarchical control system. In which case, the impact of wind power fluctuation can be absorbed by using microgrid concept with the proposed hierarchical control method
Wind farm high frequency electrical resonances : impedance-based stability analysis and mitigation techniques
No full textThis thesis was previously held under moratorium from 27th November 2019 until 27th November 2024.The installation of larger wind farms has introduced new grid integration challenges. Among these, high frequency electrical resonances caused by the cables and lines connecting the wind farm to the grid have been described in literature. These resonances may cause instability as they typically occur at frequencies close to the current controller bandwidth of the wind turbine inverter. Different mitigation techniques have been proposed in the literature to counteract these resonances. However, these techniques lack of generality as they rely on parameter tuning, which varies on a case-by-case scenario. In order to analyse the problem, an impedance-based stability approach has been applied. A systematic technique to derived the sequence-frame converter admittance
has been de�ned, and a stability study methodology including the coupling
between the positive and negative sequence converter admittances has been formulated. Compared to the existing impedance-based stability criterion, where such coupling is ignored, the proposed technique is more accurate in the system stability assessment. The study has shown that the delay introduced by the controller implementation is the main cause of the investigated wind farm stability issues. An innovative hardware implementation of the controller is proposed to compensate for this delay, without altering the converter switching frequency but making a more efficient use of the available hardware processing power. Hence, a more general and portable solution to the problem is proposed, which does not require parameter tuning. An experimental validation of the applied stability analysis methodology and of the proposed mitigation techniques has been carried out, making use of a purposed built prototype of a converter-grid system.The installation of larger wind farms has introduced new grid integration challenges. Among these, high frequency electrical resonances caused by the cables and lines connecting the wind farm to the grid have been described in literature. These resonances may cause instability as they typically occur at frequencies close to the current controller bandwidth of the wind turbine inverter. Different mitigation techniques have been proposed in the literature to counteract these resonances. However, these techniques lack of generality as they rely on parameter tuning, which varies on a case-by-case scenario. In order to analyse the problem, an impedance-based stability approach has been applied. A systematic technique to derived the sequence-frame converter admittance
has been de�ned, and a stability study methodology including the coupling
between the positive and negative sequence converter admittances has been formulated. Compared to the existing impedance-based stability criterion, where such coupling is ignored, the proposed technique is more accurate in the system stability assessment. The study has shown that the delay introduced by the controller implementation is the main cause of the investigated wind farm stability issues. An innovative hardware implementation of the controller is proposed to compensate for this delay, without altering the converter switching frequency but making a more efficient use of the available hardware processing power. Hence, a more general and portable solution to the problem is proposed, which does not require parameter tuning. An experimental validation of the applied stability analysis methodology and of the proposed mitigation techniques has been carried out, making use of a purposed built prototype of a converter-grid system
Access Repository Dashboard
Do you manage STAX (Strathclyde Repository)? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!