The Modeling of the Frequency Regulation Performance for Thermal Units and Virtual Power Plants

• Published in: 2020 5th Asia Conference on Power and Electrical Engineering (ACPEE) pp. 1940 - 1944
• Main Authors: Li, Jingbang, Gao, Ciwei, Yang, Jianlin, Chen, Songsong, Li, Xueming
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2020
• Subjects: Computational efficiency; Computational modeling; Electricity supply industry; frequency regulation performance; Linear regression; linear regression fitting; Regulation; thermal unit; Uncertainty; virtual power plant; Virtual power plants
• DOI: 10.1109/ACPEE48638.2020.9136414

Providing frequency regulation service can help generators and electric loads make a profit in electricity market. In some regulation markets, the frequency regulation performance is a key index/score and it affects the revenue of the service's provider. Some studies have focused on modeling the regulation performance score by simplifying or linearizing the performance's statistical function. However, since regulation signals are usually broadcasted every 2 seconds, there are at least 1800 variables denoting the response power in the existing models. This will make the optimal bidding in regulation market computationally expensive and difficult to solve. Moreover, the regulation revenue function is a kind of nonlinear function. It is difficult to derive optimal solution in a nonlinear optimization. To solve those problems, in this paper, we propose the models of frequency regulation performance for thermal units (TUs) and virtual power plants (VPPs) by linear regression. With the proposed models, we prove the regulation revenue functions are linear. Meanwhile, simulation illustrates that the proposed models need less variable and the models are more computationally efficient than existing methodology.