Context-Aware Stochastic Modeling of Consumer Energy Resource Aggregators in Electricity Markets

TL;DR

This work tackles uncertainty in CER aggregation for Australian NEM participation by developing three two-stage stochastic optimization frameworks—risk-neutral, robust, and chance-constrained—tailored to energy and FCAS markets while explicitly modeling BES SoC dynamics and complementarity. It advances tractability through scenario-based reformulations, affine recourse, and scalable relaxations (LP/McCormick) to handle large scenario sets, validated on real-NEM-like data with MCMC-generated forecast errors. The key contributions provide context-aware guidance for method selection, balancing feasibility, profitability, and risk across diverse operational settings, uncertainty characteristics, and decision-making preferences. Practically, the methods enable context-sensitive bidding strategies for CER aggregators, supporting ADC feasibility, revenue optimization, and scalable deployment in evolving VPP-enabled electricity markets.

Abstract

Aggregators of consumer energy resources (CERs) like rooftop solar and battery energy storage (BES) face challenges due to their inherent uncertainties. A sensible approach is to use stochastic optimization to handle such uncertainties, which can lead to infeasible problems or loss in revenues if not chosen appropriately. This paper presents three efficient two-stage stochastic optimization methods: risk-neutral, robust, and chance-constrained, to address the impact of CER uncertainties for aggregators who participate in energy and regulation services markets in the Australian National Electricity Market. Furthermore, these methods utilize the flexibility of BES, considering precise state-of-charge dynamics and complementarity constraints, aiming for scalable performance while managing uncertainty. The problems are formed as two-stage stochastic mixed-integer linear programs, with relaxations adopted for large scenario sets. The solution approach employs scenario-based methodologies and affine recourse policies to obtain tractable reformulations. These methods are evaluated across use cases reflecting diverse operational and market settings, uncertainty characteristics, and decision-making preferences, demonstrating their ability to mitigate uncertainty, enhance profitability, and provide context-aware guidance for aggregators in choosing the most appropriate stochastic optimization method.

Figures (8)

• Figure 1: (a) Market clearing prices (MCPs) for energy, regulation FCAS raise, and regulation FCAS lower, (b) Actual and forecast PV values, and 1000 PV scenario trajectories stemming from MCMC-based forecast errors (c) Actual and forecast load values, and 1000 load scenario trajectories stemming from MCMC-based forecast errors, for day 01/07/2012.
• Figure 2: Locations of the (a) houses (in blue) and (b) PV units colored by the installed capacity.
• Figure 3: State transition probability matrices for (a) PV forecast errors and (b) load forecast errors. "Yellow" indicates high probabilities, while "blue" indicates low probabilities.)
• Figure 4: Comparison of (a) profits, (b) computational times, and (c) percentage of mutual exclusivity violations of BES decisions, for different reformulations under risk-neutral setting with increasing number of scenarios.
• Figure 5: Training and test (a) profits and (b) computational times with and without recourse policies for day 01/07/2012 across different optimization methods.