Logo do repositório
 
A carregar...
Miniatura
Publicação

The Transformative Potential of Machine Learning in the Energy Distribution Sector

Utilize este identificador para referenciar este registo.
Nome:Descrição:Tamanho:Formato: 
DM-52065-2024.pdf1.03 MBAdobe PDF Ver/Abrir

Resumo(s)

Climate change poses one of the greatest challenges of the 21st century, requiring a fundamental transformation of energy systems to reduce carbon emissions and support global sustainability goals. The energy transition aims to shift away from fossil fuels towards renewable energy sources while ensuring energy security and efficiency. In this context, energy distribution companies play a crucial role in enabling this transition by integrating new technologies and managing the complexities of modern energy grids. The methodology employed in this research is a qualitative case study. Data was collected through semi-structured interviews and an analysis of external and internal documentation, providing a comprehensive understanding of the company's efforts to contribute to the energy transition, and access how Machine Learning (ML) can help challenging the current regime. This thesis examines how E-REDES, as a Distribution System Operator (DSO), is leveraging ML in order to contribute to the energy transition. Given the early stages of ML adoption in DSOs, this research acknowledges that it is still too early to derive substantial results from these initiatives, thus substantial and measurable outcomes have yet to be fully realized. Instead, the study aims to assess the current contributions of ML to the energy transition and whether these applications can drive systemic changes to the current energy regime. The findings reveal that while ML significantly improves efficiency and supports the integration of renewable energy, it primarily functions as a tool for optimization rather than disruption of the current energy regime. The study concludes that ML, in its current applications, contributes to accelerating the energy transition but does not yet have the transformative power to challenge the existing system.
As mudanças climáticas representam um dos maiores desafios do século XXI, exigindo uma transformação dos sistemas energéticos com o objetivo de reduzir as emissões de carbono e apoiar os objetivos globais de sustentabilidade. A transição energética visa reduzir a utilização de combustíveis fósseis substituindo-os por fontes renováveis de energia, assegurando, simultaneamente, a segurança e a eficiência energética. Nesse contexto, as empresas de distribuição de energia desempenham um papel crucial ao integrar novas tecnologias e gerir as complexidades das novas redes de energia. A metodologia utilizada nesta pesquisa é um estudo de caso qualitativo, com foco na E-REDES. Os dados foram recolhidos através de entrevistas semiestruturadas e da análise de documentação externa, explorando os esforços realizados pela empresa para contribuir para a transição energética. A presente dissertação pretende explorar como a E-REDES, enquanto Operadora da Rede de Distribuição (ORD), está a utilizar o Machine Learning (ML) para contribuir para a transição energética. Dado que a adoção de ML, nas DSOs, ainda se encontra numa fase inicial, este estudo reconhece que ainda não foram alcançados resultados mensuráveis e significativos. Assim sendo, o objetivo do estudo é avaliar as contribuições atuais do ML para a transição energética e se essas aplicações podem gerar mudanças sistêmicas no regime energético atual. Os resultados revelam que, embora o ML tenha potencial para melhorar significativamente a eficiência operacional e apoiar a integração de energia renovável, o mesmo funciona maioritariamente como uma ferramenta de otimização, ao invés de disrupção do atual regime energético. O estudo conclui que, considerando as suas atuais aplicações, o ML contribui para acelerar a transição energética, mas ainda não possui o potencial transformador necessário para desafiar o sistema existente.

Descrição

Master Management And Industrial Strategy

Palavras-chave

Sustainability Energy Transition Machine Learning Energy Distribution

Contexto Educativo

Citação

Santos, Beatriz Silva (2024). “The Transformative Potential of Machine Learning in the Energy Distribution Sector: A case study of e-redes”. Dissertação de Mestrado. Universidade de Lisboa. Instituto Superior de Economia e Gestão

Projetos de investigação

Unidades organizacionais

Fascículo

Editora

Instituto Superior de Economia e Gestão

Licença CC