The article presents a new framework for renewable energy scheduling, which aims to improve market efficiency and provide guidelines for private suppliers. The proposed method is called bilevel programming, which combines two levels of optimization to find the best bidding quantities for renewable energy sources (RES). This approach can help establish benchmark bidding quantities in a centralized manner, minimizing expected system cost and potentially serving as a tool for market monitoring.
Benchmark Bidding Quantities
The proposed framework focuses on the system perspective, which means that it establishes benchmark bidding quantities in a centralized way by minimizing the expected system cost. These benchmarks can serve as a reference point to guide and monitor private suppliers’ bidding strategies. By comparing the benchmark results with forecast values, a risk score can be developed to adjust how much suppliers bid in the market.
Piece-Wise Linear Constraints
If accuracy is poor for a specific problem, the authors propose using multi-segment McCormick envelopes, which introduce piece-wise linear constraints instead of linear ones. This approach allows for more flexibility in adjusting the tradeoff between accuracy and complexity. However, evaluating this strategy requires additional evaluation.
Future Extensions
The article highlights two important future extensions for this work: decomposition methods to improve computational efficiency and incorporating this framework into current electricity market designs. Decomposition methods can help speed up computations, while incorporating the framework into existing markets can enhance its practical applicability.
In conclusion, the proposed bilevel framework offers a scalable solution for renewable energy scheduling by establishing benchmark bidding quantities and potentially serving as a tool for market monitoring. By combining two levels of optimization, this method can improve market efficiency and help private suppliers adjust their bidding strategies accordingly. Future work may focus on enhancing computational efficiency or incorporating the framework into current electricity market designs to further improve its practical applicability.
