Nas últimas décadas, diversos estimadores de estado foram desenvolvidos e aplicados em sistemas elétricos de potência (SEP), dos quais se destaca o baseado no método da mínima mediana. Isso em razão desse, conhecido como estimador por mínima mediana do resíduo ponderado ao quadrado (do inglês, Weighted Least Median of Squares - WLMS), ser capaz de filtrar erros grosseiros (EGs) existentes em medidas redundantes que atraem a convergência do processo de estimação de estado. Essas medidas são chamadas de ponto de alavancamento. Todavia, o estimador WLMS requer um alto custo computacional, tornando-se inviável para aplicação, em tempo real, em SEP de grande porte. De uma forma geral, os motivos desse custo computacional são devido ao estimador WLMS exigir a realização das seguintes tarefas: (i) sorteio, dentre todas as medidas disponíveis, de diversos conjuntos observáveis de medidas com número de medidas igual ao número de variáveis de estado a serem estimadas; (ii) análise da redundância local das medidas disponíveis para cada um dos conjuntos observáveis sorteados; e (iii) execução de vários fluxos de carga. Neste trabalho, propõe-se o desenvolvimento de um estimador de estado estatisticamente robusto, baseado no método da mínima mediana, porém, viável para aplicação em tempo real em SEP de grande porte. Para isso, foram propostas novas metodologias para realização das tarefas supracitadas. A metodologia proposta para realização das tarefas (i) e (ii) faz uso da estrutura da matriz H\'delta\', pois, através dessa matriz, as análises de observabilidade e de redundância de medidas são realizadas de forma simples e direta. Para realizar a tarefa (ii), desenvolveu-se uma metodologia para cálculo de fluxo de carga, tomando por base um método de varredura direta/inversa. Por fim, para aumentar a eficiência computacional em tempo real do estimador proposto, as tarefas a serem executadas pelo mesmo que não dependem de informações, que se tornam disponíveis em tempo real, são realizadas num processo off-line. Como principais resultados deste trabalho, destacam-se: (i) um estimador de estado robusto; e (ii) uma metodologia eficiente para determinação da mínima redundância local de medidas e para sortear os conjuntos observáveis de medidas. / In the last decades several state estimators were developed and applied to power systems. Among these estimators, the one based on the least median method is of interest to us. This because the weighted least median of squares (WLMS) estimator is capable of filtering gross errors corrupting redundant measurements called leverage points. Leverage points are highly influential measurements that attract the state estimations solution towards them. However, some of the WLMS estimator tasks require excessive computing time, making that estimator impracticable to large-scale power systems in real-time environment. The WLMS estimator tasks requiring excessive computing time are: (i) selection, among all available measurements, of several samples with N non-redundant measurements for which the system is observable, where N is the number of system states to be estimated; (ii) determination of the minimum redundancy of the available measurements set; and (iii) the solution of several load flows (one for each selected samples of N measurements). This work proposes a robust state estimator based on the least median method applicable to large-scale power systems in real-time environment. In order to do this, new methodologies were proposed to perform the tasks mentioned above. The proposed methodology to perform tasks (i) and (ii) is based on the analysis of the H\'delta\' matrix structure (this analysis enables both observability and redundancy analysis in a straightforward manner). To perform task (ii), it was developed a load flow methodology based on a forward/backward sweep power flow method. Finally, in order to increase the computational efficiency of the proposed estimator in real-time environment, the tasks that do not depend on real-time information will be conducted by an off-line process. As the main results of this work we could enumerate: (i) a robust state estimator; and (ii) an efficient methodology to determine both the minimum redundancy of the available measurement set and the observable samples with N non-redundant measurements.

Nas últimas décadas, diversos estimadores de estado foram desenvolvidos e aplicados em sistemas elétricos de potência (SEP), dos quais se destaca o baseado no método da mínima mediana. Isso em razão desse, conhecido como estimador por mínima mediana do resíduo ponderado ao quadrado (do inglês, Weighted Least Median of Squares - WLMS), ser capaz de filtrar erros grosseiros (EGs) existentes em medidas redundantes que atraem a convergência do processo de estimação de estado. Essas medidas são chamadas de ponto de alavancamento. Todavia, o estimador WLMS requer um alto custo computacional, tornando-se inviável para aplicação, em tempo real, em SEP de grande porte. De uma forma geral, os motivos desse custo computacional são devido ao estimador WLMS exigir a realização das seguintes tarefas: (i) sorteio, dentre todas as medidas disponíveis, de diversos conjuntos observáveis de medidas com número de medidas igual ao número de variáveis de estado a serem estimadas; (ii) análise da redundância local das medidas disponíveis para cada um dos conjuntos observáveis sorteados; e (iii) execução de vários fluxos de carga. Neste trabalho, propõe-se o desenvolvimento de um estimador de estado estatisticamente robusto, baseado no método da mínima mediana, porém, viável para aplicação em tempo real em SEP de grande porte. Para isso, foram propostas novas metodologias para realização das tarefas supracitadas. A metodologia proposta para realização das tarefas (i) e (ii) faz uso da estrutura da matriz H\'delta\', pois, através dessa matriz, as análises de observabilidade e de redundância de medidas são realizadas de forma simples e direta. Para realizar a tarefa (ii), desenvolveu-se uma metodologia para cálculo de fluxo de carga, tomando por base um método de varredura direta/inversa. Por fim, para aumentar a eficiência computacional em tempo real do estimador proposto, as tarefas a serem executadas pelo mesmo que não dependem de informações, que se tornam disponíveis em tempo real, são realizadas num processo off-line. Como principais resultados deste trabalho, destacam-se: (i) um estimador de estado robusto; e (ii) uma metodologia eficiente para determinação da mínima redundância local de medidas e para sortear os conjuntos observáveis de medidas. / In the last decades several state estimators were developed and applied to power systems. Among these estimators, the one based on the least median method is of interest to us. This because the weighted least median of squares (WLMS) estimator is capable of filtering gross errors corrupting redundant measurements called leverage points. Leverage points are highly influential measurements that attract the state estimations solution towards them. However, some of the WLMS estimator tasks require excessive computing time, making that estimator impracticable to large-scale power systems in real-time environment. The WLMS estimator tasks requiring excessive computing time are: (i) selection, among all available measurements, of several samples with N non-redundant measurements for which the system is observable, where N is the number of system states to be estimated; (ii) determination of the minimum redundancy of the available measurements set; and (iii) the solution of several load flows (one for each selected samples of N measurements). This work proposes a robust state estimator based on the least median method applicable to large-scale power systems in real-time environment. In order to do this, new methodologies were proposed to perform the tasks mentioned above. The proposed methodology to perform tasks (i) and (ii) is based on the analysis of the H\'delta\' matrix structure (this analysis enables both observability and redundancy analysis in a straightforward manner). To perform task (ii), it was developed a load flow methodology based on a forward/backward sweep power flow method. Finally, in order to increase the computational efficiency of the proposed estimator in real-time environment, the tasks that do not depend on real-time information will be conducted by an off-line process. As the main results of this work we could enumerate: (i) a robust state estimator; and (ii) an efficient methodology to determine both the minimum redundancy of the available measurement set and the observable samples with N non-redundant measurements.

There is increasing concern about potential climate-driven regime shifts– large abrupt shifts in social-ecological systems that could have large impacts onecosystems services and human well-being. This paper aims to synthesize the potentialpathways for building resilience to such regime shifts. Ten examples from the RegimeShift Database provided the cases for analysis. Causal loop diagrams were used toanalyze feedback mechanisms at different scales and identify “leverage points” –places to intervene in the system in order to build resilience. Sixteen of these leveragepoints were identified, most of which relate to agricultural management. Mostfeedback mechanisms include at least one leverage point highlighting the potential forbuilding resilience to climate-induced regime shifts. The most common leverage pointsidentified in our analyses were vegetation cover, algae volume and atmospherictemperature. These leverage points were compared to mitigation strategies discussedby the IPCC. This comparison indicates that current climate change mitigationstrategies do not alter most of the leverage points directly. This suggests that IPCCstrategies should be broadened in order to reduce the risk of regime shifts, and theassociated impacts on human well-being.

In an ever-changing world of work Absa, as a business, is faced with various challenges including the continuous development of skills. Due to technological advancements, eLearning can provide a mechanism to rapidly build the required strategic and tactical skills that the organisation needs. This study explored the challenge of articulating the contribution of eLearning to business performance in an unbounded way. The study focused on the creation of knowledge about how the contribution of eLearning to business performance can be improved. In the process of knowledge creation, the study focused on identifying the point of value creation between Business and an eLearning intervention. This point of value creation can be seen as a leverage point. Systems Thinking was implemented as an approach in order to identify the leverage point. The following research objectives were defined: · To identify the driver problem that prevents eLearning from improving business performance. · To design the systems dynamic model that represents the driver problem. · To identify the leverage point within the systems dynamic model. · To reflect on the effect that the behaviour of the individuals, participating in the research process, has on the research inquiry. The sample of 28 focus group participants was selected from two specific divisions of Absa – the eChannels: Contact Centre and the Learning and Development Department. This sample consisted of Operational Management, Team Leaders, Contact Centre Consultants and learning design experts. Executive Management was excluded from the focus groups, but was included in the process as verifiers. This created an opportunity for Executive Management to voice their opinions. The results of the study indicate that the leverage point for successful contribution of eLearning to business performance is … A shared mental model of expectations between the participating stakeholders. Once Business and the Learning and Development Department start going through the constructive cycle of the systems dynamic model repeatedly, they will continuously build the shared mental model of expectations. This cycle will also build on the: 1) Level of visible support of the line managers; 2) Level of clarity of business needs to all relevant stakeholders; 3) Number of requests from business for eLearning opportunities; and 4) Level of awareness and understanding of appropriate eLearning interventions per target population. The effect of the positive reinforcement of the recurring cycle will ensure that eLearning continuously contributes to business performance. During the study the effect of the research process on the focus group participants as well as the effect of the focus group participants on the research process was also accounted for. Observers reflected on the behaviour of the focus group participants and found that their opinions and thought processes influenced the outcome of the study. The focus group participants felt that they had learnt something new, that the tasks set to the groups was clear and that the topics they had learnt most about were ‘systems thinking’ followed by the ’ relationship between eLearning and business performance’. Copyright / Thesis (PhD (Curriculum Studies))--University of Pretoria, 2004. / Curriculum Studies / unrestricted