Data-Driven Modelling and Operation Decision-Making of Electricity Markets with Distributed Energy Resources

Abstract

As a result of excessive carbon emissions, the consumption of fossil fuels for generating electricity can cause damage to the environment. Faced with the dilemma of energy demand and global warming, humans urgently need an economical, safe and reliable energy solution. The global power sectors are accelerating to increase the penetration of renewable energy resources to achieve the target of low carbon emissions. Compared with the conventional synchronous generators dominated electricity market, the operation of the market requires new models and decision-making strategies to accept and adapt to the high penetration of Distributed Energy Resources (DERs). To facilitate the integration of DERs, some challenging issues in the market operation need to be solved instantly from the aspects of economics and technology. In the aspect of economics, since DERs are owned by end-customers in the distribution network, DER capacity is quite small compared with power plants. End-customers are not allowed to directly participate in the bidding and clearing of the wholesale electricity market. On the other hand, considering the technology, the network constraints must be visible to the network operator, meanwhile, the DERs must be coordinated with the constraints. Also, correct operation decisions rely on accurate network topology structure and load models. To this end, some research on data-driven modeling and operation decision-making of electricity markets with DERs is proposed to solve current issues. In this thesis, a systematic data-driven framework, model and solution are established comprehensively considering the following aspects including uncertainty of market prices, visibility of distribution network topology, DERs forecasting errors, solvability and effectiveness of decision-making models, and time variation of load model characteristics.