Enhancing PV Hosting Capacity in Electricity Distribution Networks

<p dir="ltr">Surging integration rates of rooftop photovoltaic (PV) systems into electricity distribution networks play a crucial part in the decarbonisation of the energy sector and its transition toward a more sustainable system. Despite the benefits of adopting PV systems, such as reduced electricity costs for customers and lesser reliance on fossil fuels, they can cause various challenges to the electricity distribution system if suitable planning and management approaches are not adopted. Some of the adverse impacts of high PV penetration are overvoltage, revenue loss due to curtailments, increased network unbalance, and thermal overloading of feeders and transformers.</p><p dir="ltr">The hosting capacity of an electricity distribution network is defined as the maximum distributed generation that can be safely and reliably integrated without causing any adverse impacts to the grid. Undertaking a hosting capacity assessment allows distribution network service providers to plan investments more effectively and integrate more distributed energy resources into the grid, ensuring cost-effective grid expansion. Over-voltage is the most common constraint which limits the hosting capacity of distribution networks. New technologies for hosting capacity enhancement more effectively regulate the voltage of the grid, allowing more energy exports to the grid without leading to the tripping of the inverters of consumer PV systems.</p><p dir="ltr">The research work undertaken in this thesis leverages recent advancements in artificial intelligence and applies them to the domains of hosting capacity assessment, coordinated voltage control, and dynamic operating envelopes. Software tools have been developed to model and quantify the hosting capacity of electricity distribution networks in both model-based and model-free manners. Model-based methods involve the development of power flow models using the network topology and element data available to the distribution network service provider, while model-free methods utilize a smart meter data-driven, artificial intelligence-based approach to model the electricity distribution networks. Model-free methods excel in situations where topological and network element data is unavailable.</p><p dir="ltr">A coordinated voltage control scheme utilizing artificial intelligence is proposed in this thesis to enhance the hosting capacity of electricity distribution networks. Coordinated voltage control adopts a holistic approach to optimally regulate voltage, in contrast to commonly used local voltage schemes in the industry. The local voltage schemes were identified as performing sub-optimally when applied to unbalanced low-voltage distribution networks. This was investigated in detail through an analytical assessment. Furthermore, this thesis presents an artificial intelligence-based approach to quantify the enhanced real-time hosting capacity when such a coordinated voltage scheme is implemented. This approach is robust and scalable, offering faster hosting capacity estimates without the need for extensive simulations.</p><p dir="ltr">In contrast to coordinated voltage control, which requires a sophisticated communication infrastructure and assumes direct control of consumer energy systems at higher time resolutions, dynamic operating envelopes offer an indirect form of control over consumer energy systems that is less susceptible to communication breakdowns and is executed at higher time resolutions, typically between 5-30 minutes. Dynamic operating envelopes are generally defined as time- and location-specific limits on import and export, adjusted based on the available capacity of the local network or the broader power system. In the context of Australian energy systems, dynamic operating envelopes are generally evaluated and deployed by the distribution network service providers. Dynamic envelopes are a key technology that enables the transition to a more decentralized energy system, fostering the wholesale energy market participation of consumers. Such decentralized market frameworks are more competitive, and as a result, energy costs for consumers will be significantly reduced.</p><p dir="ltr">As artificial intelligence revolutionizes industries worldwide, its applications in power distribution systems remain largely untapped. This thesis proposes a novel methodology to estimate dynamic operating envelopes by leveraging artificial intelligence. The proposed approach is model-free, scalable, and robust, as proven by the numerical study undertaken, and demonstrates remarkable improvements over traditional dynamic operating envelope estimation strategies. Furthermore, this thesis presents a critical, unbiased discussion of the impact of artificial intelligence on power distribution networks and offers a fair evaluation of its advantages and pitfalls.</p><p dir="ltr">This thesis aims to illuminate new possibilities and uncover novel approaches that enhance the hosting capacity of electricity distribution networks, contributing to the achievement of Net Zero energy targets.</p>