TL;DR
This paper introduces a Power Demand-Supply Cooperative Response (PDSCR) strategy designed to maximize renewable energy utilization by synchronizing industrial demand with power supply. By using game theory to ensure fair profit distribution among participants, the strategy effectively mitigates the unpredictable fluctuations of wind power across both day-ahead and intraday time scales.
What Problem Does This Paper Address?
As we transition toward a "green" grid, the high penetration of Renewable Energy (RE)—specifically wind power—presents a massive stability challenge. Wind is notoriously fickle; it doesn't always blow when we need electricity, and it often blows hardest when demand is low.
Traditional Demand Response (DR) programs encourage consumers to shift their usage, but these are often centralized and top-down. In decentralized markets, participants (like industrial factories and utility companies) act in their own self-interest. If a factory is asked to ramp up production to soak up excess wind power, but the financial incentive doesn't cover their increased operational costs, they simply won't do it. This "conflict of interest" leads to wasted renewable energy (curtailment) and higher system costs. The authors tackle the question: How can we coordinate supply and demand in a way that everyone wins and renewable energy isn't wasted?
Methods
The authors propose a two-stage strategy called PDSCR:
- Day-Ahead (Centralized): Using a Bi-level Multi-Objective (BMO) optimization model based on Stackelberg game theory. Here, the utility center acts as the "leader" (setting prices and thermal unit schedules), and Industrial Production Enterprises (IPEs) act as "followers" (optimizing their production schedules based on those prices).
- Intraday (Decentralized & Cooperative): To handle real-time wind fluctuations, the authors move from a leader-follower model to a cooperative game. They extend the Nash Bargaining Solution (NBS) to handle multi-individual, nonlinear utilities.
The "secret sauce" of the paper is the development of Multi-Individual Profit Distribution Marginal Solutions (MIPDMSs). These are mathematical frameworks that ensure the "cooperative profit" (the money saved by working together) is distributed fairly. They explore two flavors:
- Equal Distribution: Everyone gets an equal slice of the savings.
- Contribution-Based Distribution: Participants who do more of the "heavy lifting" (adjusting their power the most) get a larger share.
Datasets
The strategy was validated using two primary testbeds:
- Modified IEEE 24-bus system: A standard benchmark in power systems research, modified to include two wind farms and a Parallel Manufacturing Process (PMP) system representing industrial load.
- Actual Yantai 26-bus system: A real-world power system model from China, providing a practical "stress test" for the proposed algorithms.
Key Insights and Ideas
- Cooperation over Competition: The paper proves that a profit-driven cooperative model is more effective at reducing wind curtailment than independent, self-interested dispatch.
- The Fairness Factor: The authors demonstrate that "unreasonable" profit distribution is a primary barrier to grid flexibility. If the utility center keeps all the savings, industrial loads have no motive to help balance the grid.
- Contribution Matters: While equal profit sharing is "fair" in a basic sense, the contribution-based MIPDMS was found to be more effective. It expands the "feasible region" of profit utility, meaning it creates a stronger incentive for participants to respond aggressively to wind fluctuations.
- Bi-Level Synchronization: By linking long-term (day-ahead) thermal unit commitment with short-term (intraday) industrial flexibility, the system creates a "buffer" that makes high-renewable penetration much safer for the grid.
Key Results
The case studies yielded impressive quantitative improvements over standard day-ahead strategies (Case 1):
- Wind Curtailment Reduction: In the Yantai system, the cooperative intraday strategy (Case 3) reduced wind curtailment from 3.61 × 10³ MW to 1.86 × 10³ MW—nearly a 50% improvement.
- Cost Savings: Total system costs on the demand-supply side dropped significantly. In one representative scenario, costs fell from $5.42 × 10⁵** to **$4.48 × 10⁵.
- Pareto Efficiency: The use of Multi-Individual Optimal Marginal Models (MIOMM) ensured that the system reached a state where no participant could be made better off without making another worse off, effectively solving the conflict of interest.
Limitations and Future Work
While the results are promising, the authors acknowledge several hurdles for real-world implementation:
- Energy Storage: The current model does not yet integrate large-scale battery energy storage systems, which would add another layer of complexity and flexibility.
- Communication Infrastructure: The decentralized cooperative response relies on precise, real-time data exchange. Any latency or "noise" in the communication network could degrade the strategy's effectiveness.
- Cybersecurity: As the grid becomes more decentralized and "talkative," it becomes more vulnerable to cyber-attacks. The authors suggest that future research must incorporate "cyber-resiliency" into these cooperative models.
Summary updated