Optimizing Flexible Resource Bidding in Nordic Ancillary Service Markets under Distributional Robustness and P90 Requirement

Core Concepts

An aggregator can leverage the P90 requirement in Nordic ancillary service markets to optimize its bidding strategy for a portfolio of stochastic flexible resources, such as electric vehicles, using a distributionally robust joint chance-constrained optimization model. This approach addresses the challenge of non-stationarity in the uncertain baseline power consumption.

Abstract

The paper presents a framework for an aggregator of stochastic flexible resources, such as electric vehicles, to optimize its bidding strategy in Nordic ancillary service markets under the P90 requirement. The key insights are:

The P90 requirement, introduced by the Danish TSO Energinet, allows stochastic flexible resources to participate in ancillary service markets by accepting a 10% probability of reserve shortfall. This requirement can be formulated as a joint chance-constrained optimization problem for the aggregator.

To address potential non-stationarity in the uncertain baseline power consumption, the authors propose a distributionally robust formulation of the joint chance-constrained problem. This allows the aggregator to consider a set of possible distributions for the baseline power, rather than relying on a single empirical distribution.

The authors also investigate the perspective of the TSO, who can adjust the P90 requirement and the level of conservativeness required from the aggregators. This is formulated as a bi-level optimization problem, where the TSO aims to maximize the total reserve capacity procurement while minimizing the expected reserve shortfall.

Numerical results using a simulated case study of an electric vehicle aggregator demonstrate the importance of distributional robustness in non-stationary environments. The authors show how the aggregator can leverage the P90 requirement to its advantage by optimizing its bidding strategy, while the TSO can find the right balance between increased flexibility supply and uncertainty of delivery.

Stats

The portfolio of 20 electric vehicles has a stochastic baseline power consumption that exhibits a positive drift over time, representing the effect of colder weather leading to longer charging periods.

Quotes

"Leveraging this requirement, we develop a distributionally robust joint chance-constrained optimization model for aggregators of flexible resources to optimize their volume of reserve capacity to be offered."
"We show how distributional robustness is key for the aggregator when making bidding decisions in a non-stationary uncertain environment."

Key Insights Distilled From

Leveraging P90 Requirement: Flexible Resources Bidding in Nordic Ancillary Service Markets

by Peter A.V. G... at arxiv.org 04-22-2024

https://arxiv.org/pdf/2404.12807.pdf

Deeper Inquiries

How can the proposed framework be extended to consider a heterogeneous portfolio of stochastic flexible resources, such as a mix of electric vehicles, heat pumps, and renewable energy sources

To extend the proposed framework to consider a heterogeneous portfolio of stochastic flexible resources, including a mix of electric vehicles (EVs), heat pumps, and renewable energy sources, several adjustments and enhancements can be made. Firstly, the empirical distribution used to represent the uncertain baseline power consumption of each resource type would need to be expanded to incorporate the unique characteristics and variability of each technology. This would involve collecting historical data and constructing separate empirical distributions for EVs, heat pumps, and renewable energy sources.
Secondly, the ambiguity set defined by the Wasserstein distance could be modified to account for the diverse nature of the portfolio. By creating distinct ambiguity sets for each resource type within the portfolio, the framework can capture the variability and uncertainty associated with different technologies. This would enable the aggregator to make more informed and robust bidding decisions, considering the specific characteristics of each resource.
Furthermore, the optimization model, such as the Distributionally Robust Joint Chance-Constrained Program (DRJCCP), would need to be adapted to accommodate the heterogeneous portfolio. This may involve introducing additional decision variables, constraints, and objective functions tailored to the unique attributes of EVs, heat pumps, and renewable energy sources. By customizing the model for each resource type, the aggregator can optimize its bidding strategy effectively while ensuring reliability and profitability across the entire portfolio.
In summary, extending the framework to consider a heterogeneous portfolio of stochastic flexible resources requires customizing the empirical distributions, ambiguity sets, and optimization model to account for the diverse characteristics and uncertainties associated with different technologies. By incorporating these adjustments, the aggregator can enhance its bidding strategy and maximize the value of the portfolio in Nordic ancillary service markets.

What are the potential implications of the aggregator's bidding strategy on the overall electricity market dynamics, including energy prices and system stability

The aggregator's bidding strategy can have significant implications on the overall electricity market dynamics, influencing energy prices and system stability in several ways.

Energy Prices: The aggregator's bidding decisions, particularly in response to the P90 requirement and conservativeness levels, can impact the clearing prices in the ancillary service markets. By offering reserve capacity with varying levels of reliability and conservativeness, the aggregator may influence the supply-demand balance and price formation in these markets. Higher conservativeness levels, leading to lower reserve capacity bids, could potentially drive up prices due to increased scarcity of reliable resources. Conversely, more aggressive bidding strategies may result in lower prices but with higher risks of reserve shortfall.

System Stability: The aggregator's participation in ancillary service markets and its bidding strategy play a crucial role in maintaining system stability. By offering flexible resources for balancing services, the aggregator contributes to grid reliability and frequency regulation. However, the impact of the aggregator's bids on system stability depends on the accuracy of its forecasting, the reliability of its resources, and the level of reserve capacity procured. Inaccurate bids or excessive reliance on stochastic resources without adequate backup could introduce volatility and uncertainty, potentially affecting system stability.

Market Competition: The aggregator's bidding behavior can also influence market competition and the overall market structure. By optimizing its bidding strategy based on the P90 requirement and conservativeness levels, the aggregator may gain a competitive advantage in the ancillary service markets. This could impact the market dynamics, competitiveness, and participation of other market players, potentially leading to changes in market concentration and efficiency.

In conclusion, the aggregator's bidding strategy has far-reaching implications on energy prices, system stability, and market dynamics in the electricity market. By carefully considering the P90 requirement, conservativeness levels, and the reliability of its resources, the aggregator can navigate these dynamics effectively and contribute to a more efficient and stable electricity market.

How can the TSO's objective function be further refined to incorporate the financial incentives and penalties associated with reserve capacity procurement and shortfall, respectively

To refine the Transmission System Operator's (TSO) objective function and incorporate financial incentives and penalties associated with reserve capacity procurement and shortfall, the following enhancements can be considered:

Cost Minimization: Instead of solely focusing on maximizing reserve capacity procurement, the TSO can modify the objective function to minimize the total cost of reserve procurement. This cost optimization would include the expenses associated with securing reserve capacity from aggregators, penalty costs for reserve shortfall, and any financial incentives offered to encourage reliable bidding behavior. By balancing the cost of procurement with the penalties for non-compliance, the TSO can achieve a more cost-effective and efficient reserve capacity management strategy.

Penalty Structure: Introducing a penalty structure within the objective function would incentivize aggregators to adhere to the P90 requirement and conservativeness levels. By assigning penalties for reserve shortfall based on the degree of violation and the impact on system stability, the TSO can encourage aggregators to bid more reliably and responsibly. These penalties can be dynamically adjusted based on market conditions, the severity of the shortfall, and the aggregator's historical performance.

Incentive Mechanisms: In addition to penalties, the TSO can incorporate incentive mechanisms into the objective function to reward aggregators for reliable and accurate bidding. Financial incentives, such as bonus payments or performance-based rewards, can motivate aggregators to offer reserve capacity with higher reliability and lower uncertainty. By aligning the incentives with the TSO's reliability and cost objectives, the aggregator's behavior can be steered towards more responsible and efficient bidding practices.

By refining the TSO's objective function to include cost minimization, penalty structures, and incentive mechanisms, the TSO can create a more robust and effective framework for managing reserve capacity procurement and ensuring system stability. These enhancements would promote responsible bidding behavior, improve market efficiency, and enhance the overall reliability of the electricity system.