Predicting and Publishing Accurate Imbalance Prices Using Monte Carlo Tree Search

TL;DR

The paper tackles the challenge of publishing accurate real-time imbalance prices in a highly volatile, RES-driven grid by introducing a model-based planning framework that combines an NN NRV forecaster with a cluster of RL-controlled virtual batteries and Monte Carlo Tree Search. The approach explicitly accounts for implicit BRP responses, enabling the TSO to publish prices that closely approximate the final settlement price. Key findings show substantial price-forecast improvements (up to 20.4% MAE reduction under ideal conditions and 12.8% under realistic conditions) and insights into multi-objective trade-offs between price accuracy, NRV reduction, and balancing costs. The work demonstrates the feasibility and value of an MCTS-based imbalance price publication tool, offering a path toward more proactive and robust TSO-enabled demand response schemes with practical implications for grid stability and market participation.

Abstract

The growing reliance on renewable energy sources, particularly solar and wind, has introduced challenges due to their uncontrollable production. This complicates maintaining the electrical grid balance, prompting some transmission system operators in Western Europe to implement imbalance tariffs that penalize unsustainable power deviations. These tariffs create an implicit demand response framework to mitigate grid instability. Yet, several challenges limit active participation. In Belgium, for example, imbalance prices are only calculated at the end of each 15-minute settlement period, creating high risk due to price uncertainty. This risk is further amplified by the inherent volatility of imbalance prices, discouraging participation. Although transmission system operators provide minute-based price predictions, the system imbalance volatility makes accurate price predictions challenging to obtain and requires sophisticated techniques. Moreover, publishing price estimates can prompt participants to adjust their schedules, potentially affecting the system balance and the final price, adding further complexity. To address these challenges, we propose a Monte Carlo Tree Search method that publishes accurate imbalance prices while accounting for potential response actions. Our approach models the system dynamics using a neural network forecaster and a cluster of virtual batteries controlled by reinforcement learning agents. Compared to Belgium's current publication method, our technique improves price accuracy by 20.4% under ideal conditions and by 12.8% in more realistic scenarios. This research addresses an unexplored, yet crucial problem, positioning this paper as a pioneering work in analyzing the potential of more advanced imbalance price publishing techniques.

Figures (17)

• Figure 1: Influence cycle between the BRPs and the TSO. The TSO publishes an approximation of the imbalance price, triggering an implicit reaction from the BRPs. The reaction influences the grid's SI, hence impacting the final imbalance price at the closure of the settlement period.
• Figure 2: Sample bid ladder, generated using Belgian data from 2023.
• Figure 3: Timeline of the major energy markets and services, and the time position of where our algorithm would work.
• Figure 4: Example showing a possible progression of the imbalance prices following a change in sign of the system imbalance. The imbalance price formula applied every minute of the quarter hour does not accurately predict the final price when a change in sign of the system imbalance occurs close to the end of the quarter.
• Figure 5: Historical analysis of the applied imbalance prices (left) and of the system imbalances (right) in Belgium, showing the mean value (line) and different quantile values (colored bands) up to the 1%-99% quantile interval. In the last few years, the imbalance prices have been facing a remarkable increase in magnitude. This seems to be independent of the SI deviations, which have been mostly stable for the last decade. This increment in imbalance price magnitudes creates a big opportunity for BRPs that wish to exploit the mechanism. As this trend continues, we can then expect the imbalance implicit responses to grow.