Bidding and Dispatch Strategies for Electric Vehicle Aggregators in Joint Energy and Regulation Markets: Quantifying, Pricing, and Optimizing EV Flexibility

Core Concepts

This paper proposes a novel real-time bidding and dispatch strategy for Electric Vehicle Aggregators (EVAs) participating in joint energy and regulation markets, emphasizing the quantification, pricing, and optimal utilization of EV flexibility while considering both EVA profitability and EV user preferences.

Abstract

• Bibliographic Information: Xu, M., Guo, Y., & Sun, H. (2024). Bidding and Dispatch Strategies with Flexibility Quantification and Pricing for Electric Vehicle Aggregator in Joint Energy-Regulation Market [Preprint]. arXiv:2411.02089.

• Research Objective: This paper aims to develop an online bidding and dispatch model for EVAs in a joint energy-regulation market, addressing the challenges of quantifying and pricing EV flexibility, managing uncertainties from electricity markets and EV behaviors, and ensuring fair and feasible power dispatch to individual EVs.

• Methodology: The authors propose a three-layer framework with the ISO, EVA, and EVs. They introduce a method for quantifying EV flexibility as a tradable commodity, allowing EVAs to set flexibility prices based on bid-in supply curves from EV users. A stochastic MPC-based bidding model is formulated to minimize the EVA's total cost while considering EV charging demand and market uncertainties. An optimal power dispatch protocol is proposed, ensuring EVA profitability and fairness to EV users. An affine mapping control strategy based on parametric linear programming is employed for fast online dispatch decisions.

• Key Findings: The proposed framework effectively quantifies and prices EV flexibility, enabling EVAs to procure flexibility from EV users based on their preferences. The stochastic MPC-based bidding model allows for real-time bidding decisions considering market uncertainties. The affine mapping control strategy enables fast and efficient online power dispatch to individual EVs in response to regulation signals.

• Main Conclusions: The proposed scheme demonstrates the effectiveness of incorporating EV flexibility quantification and pricing into EVA bidding and dispatch strategies. The solution methodology proves to be computationally efficient and applicable in real-time, enhancing both EVA profitability and power system reliability.

• Significance: This research contributes significantly to the field of EV aggregator optimization by addressing the critical challenges of EV flexibility management in electricity markets. The proposed framework provides a practical and efficient solution for EVAs to participate in ancillary services, promoting the integration of EVs into the smart grid.

• Limitations and Future Research: The paper assumes the EVA operates as a price-taker in the electricity market. Future research could explore scenarios where the EVA acts as a price-maker. Additionally, the study focuses on the PJM market structure; further investigation is needed to adapt the proposed framework to other market designs.

Stats

EVs have a battery capacity of 50 kWh.
The SoC ranges from a minimum of 20% to a maximum of 90%.
Charging efficiency (ηc) is 90%, and discharging efficiency (ηd) is 93%.
The simulation uses historical data from PJM in April 2023 for energy and regulation market prices.
Extreme frequency regulation scenarios and typical scenarios are defined, with probabilities derived from historical PJM RegD data in April 2020.
The parameter space for frequency signals is divided into intervals of 0.1 to capture uncertainty.
EVs are categorized into four types based on flexibility preference: conservative (αn = 0.2), cautious (αn = 0.6), proactive (αn = 1), and risky (αn = 1.4).

Quotes

Key Insights Distilled From

Bidding and Dispatch Strategies with Flexibility Quantification and Pricing for Electric Vehicle Aggregator in Joint Energy-Regulation Market

by Manqi Xu, Ye... at arxiv.org 11-05-2024

https://arxiv.org/pdf/2411.02089.pdf

Deeper Inquiries

How might the increasing prevalence of renewable energy sources impact the bidding strategies of EVAs in the future?

The increasing prevalence of renewable energy sources (RES) like solar and wind power will significantly impact the bidding strategies of EV aggregators (EVAs) in several ways:

Increased Volatility and Arbitrage Opportunities: RES, being intermittent and unpredictable, introduce volatility into electricity prices. EVAs can leverage this volatility by charging EVs when energy prices are low (typically during high RES generation periods) and discharging or reducing charging when prices are high. This energy arbitrage opportunity will become a key aspect of EVA bidding strategies.
Greater Emphasis on Flexibility: As RES penetration increases, the need for grid flexibility to balance supply and demand fluctuations also rises. EVAs, with their ability to adjust EV charging/discharging patterns, become valuable providers of this flexibility. Bidding strategies will need to prioritize and accurately quantify this flexibility to capitalize on ancillary service markets like frequency regulation, as highlighted in the paper.
Dynamic Pricing Signals: Future electricity markets might adopt dynamic pricing schemes that reflect the real-time availability of RES. EVAs will need to adapt their bidding strategies to respond to these dynamic price signals, potentially using advanced forecasting techniques and optimization algorithms to maximize profits.
Coordinated RES-EV Charging: EVAs could evolve to manage both EV charging and RES generation, optimizing the combined system to provide grid services. This would require sophisticated bidding strategies that consider both the stochastic nature of RES and the flexibility potential of EVs.
In essence, the rise of RES necessitates a shift towards more dynamic and flexible bidding strategies for EVAs, focusing on exploiting price volatility, providing grid services, and potentially integrating with RES management systems.

Could the proposed framework be adapted to accommodate heterogeneous EV battery characteristics and charging capabilities within an EVA's fleet?

Yes, the proposed framework can be adapted to accommodate heterogeneous EV battery characteristics and charging capabilities within an EVA's fleet. Here's how:

Individual EV Constraints: The framework already incorporates individual EV constraints like battery capacity (Bn), minimum and maximum state of charge (SoC) limits (Smin
n, Smax
n), and maximum charging/discharging power (pmax/min
n). These parameters can be individually specified for each EV in the fleet, capturing the heterogeneity in battery characteristics.
Flexibility Quantification: The flexibility quantification method (Equation 5) considers individual EV discharging efficiency (ηd
n) and power boundaries (p+/−
n,t), allowing for variations in charging/discharging capabilities across different EV models.
Flexibility Supply Curves: Each EV submits its own flexibility supply curve (Equation 7), reflecting its unique willingness to provide flexibility based on its battery characteristics, charging needs, and user preferences. This allows the EVA to aggregate diverse flexibility offerings from the fleet.
Dispatch Optimization: The power dispatch problem (P2) already considers individual EV power limits and flexibility commitments (∆pup/dn
n,t) when determining the optimal charging/discharging power for each EV. This ensures that the dispatch strategy respects the heterogeneous capabilities of the fleet.
Therefore, by incorporating individual EV parameters and preferences into the existing framework, the EVA can effectively manage a fleet with diverse battery characteristics and charging capabilities. This adaptation ensures fair compensation for flexibility, optimal utilization of EV resources, and reliable grid service provision.

What are the potential privacy concerns associated with sharing EV charging data with an aggregator, and how can these concerns be addressed within this framework?

Sharing EV charging data with an aggregator raises valid privacy concerns, primarily related to:

Location Tracking: Charging data, especially if time-stamped, can reveal an EV user's location and movement patterns, potentially exposing sensitive information about their daily routines and frequented places.
Charging Habits: Detailed charging data can reveal an individual's charging habits, including frequency, duration, and preferred times. This information could be used to infer personal preferences or even predict future behavior.
Data Security: Sharing data with an aggregator introduces risks related to data breaches and unauthorized access. If the aggregator's systems are compromised, sensitive EV charging data could be exposed.
Addressing these privacy concerns within the framework requires a multi-faceted approach:

Data Minimization: The framework should only require EV users to share the minimum amount of data necessary for participation. Instead of raw, time-stamped charging data, aggregated and anonymized information could be shared, preserving individual privacy.
Data Aggregation and Anonymization: The EVA can aggregate data from multiple EVs before using it for bidding or dispatch decisions. This aggregation, combined with anonymization techniques that remove personally identifiable information, can help protect individual user privacy.
Secure Communication and Storage: Implementing robust security measures, such as encryption for data transmission and secure storage solutions, is crucial to prevent unauthorized access and data breaches.
Transparency and Control: EV users should be informed about what data is collected, how it is used, and for what purpose. Providing users with control over their data, including the option to opt-out of sharing certain information, is essential for building trust.
Privacy-Preserving Technologies: Exploring privacy-preserving technologies like differential privacy or federated learning can enable the EVA to perform computations and optimize bidding strategies without accessing raw, individual-level data.
By incorporating these privacy-enhancing measures, the framework can mitigate privacy concerns associated with EV charging data sharing. This fosters user trust, encourages wider participation in EV aggregation programs, and unlocks the full potential of EVs for grid support while safeguarding user privacy.