A multi-stage data-driven IGDT-RO model with chance compensation for optimizing bidding of RES aggregator in competitive electricity markets.

Driven by environmental policies and cost reduction efforts, renewable energy sources (RESs) become increasingly popular worldwide. It is a promising way to integrate dispersed RESs in the form of aggregators into the electricity market. This paper focused on collaborative bidding for an aggregator that integrates wind, solar, hydropower and energy storage system (ESS) in day-ahead (DA) and intraday (ID) markets. We propose a comprehensive data-driven based information gap theory-robust (DIGDT-RO) to handle the multi-stage optimal bidding for the RES aggregator. The RO approach is presented to model the uncertainty of ID electricity price, while uncertainties related to wind and solar generation are considered in DIGDT, which allows the aggregator to adopt risk-averse or risk-seeking strategies towards generation fluctuations. In DIGDT, the forecasted error of wind and solar is estimated by a novel confidence interval-based ambiguity set construction method (CIAS), and then the possibility of hydropower and ESS compensating for power deviation is modeled by chance constraints. The numerical results verify the good profitability and superior adaptability of the proposed model towards uncertainties. • A multi-stage bidding model for the RES aggregator is developed. • A novel DIGDT model with chance compensation of hydropower and storage is proposed. • The characteristics are constructed by a confidence interval-based ambiguity set method (CIAS). • The coupled risk of DA and ID markets is managed. • The flexibility of controllable generators is fully exploited. [ABSTRACT FROM AUTHOR]