Abstract:
Conventional electricity network and market are experiencing increasing penetration of distributed generation (DG). The primary driving forces come from environmental concerns and economic benefits. Distributively attached with existing power networks, renewable energy resources bring much complexity and uncertainty into the traditional producer-consumer system. Additionally, existing power network’s centralised control structure has demonstrated its inefficiency with the requirement of massive computational power, the inefficiency of expansion and the degraded fault-tolerance. These challenges have brought opportunities for new technologies such as distributed control, multi-agent system (MAS) and Blockchain. This thesis intends to address some of the challenges and equip conventional distribution system and electricity market with better resiliency, flexibility and efficiency to handle increased localised peer-to-peer transactions. Firstly, a blockchain-based distributed transactive energy market is proposed. Participants in the market are abstracted as intelligent virtual agents and actively performing transactions through smart contracts. A market trading mechanism that utilises a distributed supply function bidding is also proposed. The trading mechanism has been implemented in two di↵erent scenarios. Within the first scenario, market participant is actively seeking optimal benefit by solely considering its own status. We proved the existence of a competitive equilibrium state where all agents gain their maximised benefit, and the market gains its optimal social welfare. In the second scenario, market participants are non-cooperative and have access to competitors’ bids. By strategically placing bids or prices, each market trader is maximising its own benefit. It is proved that there exists a Nash equilibrium where the system’s social welfare is optimised, and each participant’s benefit is simultaneously met. A distributed market pricing scheme is also proposed to help the market converge to the equilibrium state iteratively. Moving from market framework, this thesis shifts its focus to the internal structure and components of market agents. A versatile prosumer energy management (PEM) module that can represent di↵erent electricity market participants is designed. Thanks to the intelligent PEM mechanism, it is able to manage the consumption and generation from subordinate components, calculate power surplus or deficit, seamlessly switch role between Buyer, Seller and Non-Trader. By charging and discharging energy storage devices, the PEM module helps market agents to trade power surplus at the time that suits their benefit. Additionally, our simulation demonstrates the merit of the PEM module as the target household is able to update its role based on the current condition of generation and consumption, and has gained more self-sufficiency. Finally, this research is extended further to the control mechanism of PEM’s subordinate distributed generator. A control architecture of an islanded microgrid hosting blockchain-based transactive energy market is proposed. Within the control framework, DGs are acting as intelligent agents and continuously exchanging information so that system stability can be gained iteratively. A cooperative frequency controller is also proposed to restore DG’s output frequency back to nominal and synchronise it with other DGs’ outputs. Thanks to the shared information, frequency within the whole microgrid can be stabilised. This thesis follows a top-down approach from distributed transactive energy market framework design to prosumer energy management mechanism and co-operative control of distributed generators. It considers existing and up-coming challenges and improves current distribution network by combining technologies such as MAS, Blockchain, distributed control and Game Theory.