Intelligent Bidding Strategies in Local Electricity Markets: A Simulation-based Analysis

The integration of PV-generated electricity from households and commercial buildings into the electricity mix offers opportunity to reduce the greenhouse gas emissions released in the Earth's atmosphere. The rising quantity of these distributed energy resources (DER) and their fluctuating energy production makes it challenging for the grid operators to manage the grids. This in turn leads to an increase in retail electricity prices due to the rising grid fee. Local electricity markets (LEMs) have proven to be an effective and efficient tool to manage the local generated electricity and ensure that electricity is traded and consumed close to its point of production. The market design factors influencing the performance of an LEM are analysed using a decentralized autonomous area agent (D3A) simulation framework. The market design factors investigated are prosumers-to-consumers ratio (nPC), pricing scenarios and bidding strategies. The results were compared using performance indicators such as self sufficiency, share of individual savings (SIS) of the participants, share of market savings (SMS) and average buying rate. The simulation results show that the performance of an LEM on addition of new participant depends upon the type of participants added to the market. Furthermore, using intelligent bidding strategy like Q-learning increases the self sufficiency of the local community close to their threshold value without the addition of flexibility or storage options.