Abstract
As power grids integrate increasing levels of intermittent renewable energy, the need for flexible demand-side resources becomes critical. While intermediate spot markets offer a theoretically efficient way to trade Demand Response (DR), they face significant regulatory and infrastructure hurdles. This paper proposes a novel over-the-counter (OTC) options market for DR, allowing Load Serving Entities (LSEs) to hedge against real-time uncertainty through bilateral call options. Using stochastic optimization and empirical data from CAISO, the authors demonstrate that a well-designed options market can achieve near-optimal economic efficiency while bypassing the complexities of centralized market oversight.
Introduction
Modern wholesale electricity markets typically rely on a two-settlement system: a day-ahead market for bulk energy procurement and a real-time market to manage immediate imbalances. However, the rapid growth of variable energy resources (VERs), such as wind and solar, has introduced significant forecast errors that complicate real-time balancing.
Demand Response (DR)—the practice of incentivizing consumers to reduce electricity usage during peak periods—is a zero-emission alternative to traditional peaker plants. Despite its potential, the integration of DR into the grid is hindered by the lack of efficient trading mechanisms. While an intermediate spot market (operating between day-ahead and real-time) could theoretically optimize DR dispatch, the practical barriers—including price volatility, high infrastructure costs, and the need for extensive regulatory approval—make it difficult to implement. This research explores whether a simpler, bilateral options market can provide the same benefits with significantly less overhead.
Market Design and Methodology
The authors evaluate the energy scheduling process through a multi-stage stochastic optimization framework, comparing three distinct market structures:
- The Social Planner Benchmark: A theoretical ideal where a single entity possesses perfect information and optimizes all stages (day-ahead, intermediate, and real-time) to minimize total system costs.
- Intermediate Spot Market: A market utilizing "contingent pricing" based on updated wind forecasts. While mathematically proven to reach the social optimum, this model is difficult to implement due to its requirement for centralized clearing and its exposure of participants to high price volatility.
- The Options Market: The core contribution of this paper is an OTC market where LSEs purchase call options from DR aggregators. In this model, the LSE pays an upfront option price for the right to curtail load later. If the LSE exercises this right, they pay a pre-negotiated strike price.
To ensure market fairness and prevent the LSE from exercising monopsony power, the authors also propose a "redesigned" symmetric options market. In this configuration, the aggregator bundles DR options directly with the LSE’s day-ahead energy purchases, creating a more balanced incentive structure.
Empirical Validation
The proposed models were validated using numerical simulations based on real-world data:
- Wind Generation: 5-minute resolution data from the California Independent System Operator (CAISO) (Nov 2019 – May 2020).
- Pricing: Day-ahead and real-time Locational Marginal Prices (LMPs) from Berkeley, California (Jan 2020).
The simulation focused on a 3MW load scenario to compare the total system costs across different market designs.
Key Insights and Results
The research yields several critical findings regarding the viability of options for grid management:
- Stability for Providers: Unlike spot markets, where revenue is uncertain and volatile, the options market provides DR aggregators with guaranteed upfront income via the option price. This stability is crucial for justifying capital investments in DR technology.
- Regulatory Ease: Because these are private, bilateral OTC contracts, they do not require the same level of regulatory oversight or centralized infrastructure as a formal spot market tier.
- Efficiency Parity: The authors prove that the options market can achieve a competitive equilibrium. Most importantly, by setting an optimal strike price, the options market can nearly match the efficiency of the social planner.
Quantitative Performance
In the 3MW simulation, the system costs were as follows:
- Baseline (No DR): $55.469
- Ideal Spot Market (Social Optimum): $52.127
- Redesigned Options Market: $52.154
The results indicate that the options market captures nearly 100% of the economic benefits of an idealized spot market, with a negligible cost difference of only ~$0.03 in the scaled model.
Conclusion and Future Work
The proposed options market offers a pragmatic path forward for scaling Demand Response. It provides the economic efficiency of a complex spot market while maintaining the simplicity of private contracts.
However, the authors note a significant limitation: achieving near-perfect efficiency requires the system operator to calculate an "optimal" strike price, which necessitates detailed knowledge of the participants' utility functions—information that is often proprietary or difficult to obtain. Future research will focus on expanding this model to include multiple intermediate stages and investigating the impact of market power in scenarios where a single LSE dominates the local grid.
Summary updated