Optimal participation of energy communities in electricity markets under uncertainty. A multi-stage stochastic programming approach

TL;DR

The paper tackles how energy communities can optimally participate in day-ahead, reserve, and intraday electricity markets under uncertainty from renewable generation and prices. It introduces a novel $34$-stage multi-stage stochastic programming model with scenario-tree generation and scenario reduction, enabling combined buying-and-selling bids and co-optimization of aggregated flexible demand alongside solar and battery storage. A December 2023 case study using real data demonstrates that energy communities can achieve positive social welfare on about $75\%$ of days, with reserve markets yielding substantial income and day-ahead participation incurring costs, while intraday actions help adjust positions. The work has practical regulatory relevance for European markets and provides a framework adaptable to aggregators or large-scale storage owners, highlighting the value of stochastic planning in energy-community operation and market integration.

Abstract

We propose a multi-stage stochastic programming model for the optimal participation of energy communities in electricity markets. The multi-stage aspect captures the different times at which variable renewable generation and electricity prices are observed. This results in large-scale optimization problem instances containing large scenario trees with 34 stages, to which scenario reduction techniques are applied. Case studies with real data are discussed to analyse proposed regulatory frameworks in Europe. The added value of considering stochasticity is also analysed.

Figures (6)

• Figure 1: Timeline of the bid submission windows and market clearing processes of the markets considered.
• Figure 2: Day-ahead buying (left, blue) and selling (right, red) bid examples. The points in orange represent the quantity-price pairs $\{(q_j, p_j) \ | \ j \in \mathcal{J} \}$ that constitute the bid.
• Figure 3: Combined buying and selling bidding curve.
• Figure 4: Behaviour summary of the energy community in a scenario during a whole day. Day-ahead position (DA, dark red), aggregated intraday position (IM, orange), imbalances (IB, black line), flexible demand consumption (dark green), battery behaviour (bright green), wind generation (light blue), PV generation (light green).
• Figure 5: Percentiles of PV generation [MWh] (top left), wind generation [MWh] (top right), flexible electricity demand [MWh] (bottom left) and battery's state of charge [MWh] (bottom right) over a day. In every plot, there is: maximum and minimum (dashed black line), 10-90 percentile (light blue), 25-75 percentile (dark blue) and median (black line).