A hybrid stochastic-interval Mean–CVaR model for the wind-storage system offering strategy under uncertainties

As an integral component of the green-oriented transition of energy, wind-storage systems have experienced vigorous development in recent years. The integration of wind-storage systems into the day-ahead market (DAM) is challenged by the multiple uncertainties associated with wind power and electricity prices. This significantly impacts the design of the offering strategy for wind-storage system operators and creates substantial financial risks for their operations. To address this issue, this paper proposes a hybrid stochastic-interval Mean–CVaR (MCVaR) model for designing flexible and diverse offering strategies that cater to different risk appetites. The model employs scenario generation to depict internal uncertainty and utilizes interval formulation to model external uncertainty. Firstly, a three-layer optimal model for scenario generation is proposed to tackle the internal uncertainty commonly associated with wind power. The scenario generation is aligned with the physical information model. Then, a virtual decision pre-construction and evaluation model is established to minimize the decision error. The corresponding solution process is designed accordingly. Secondly, an stochastic-interval MCVaR model is developed to account for the interval uncertainty of electricity price and the risk appetite of wind-storage system operators. In conjunction with the wind power scenarios, the offering strategy of the wind-storage system is designed based on this model. Finally, the case study verifies the superiority of the hybrid stochastic-interval optimization framework. The impact of interval uncertainty on the offering strategy and income of wind-storage systems is analyzed in detail, aiming to provide a reliable economic benefit analysis. Multiple offering strategies are developed for operators of wind-storage systems based on different risk appetites. • The three-layer model of scenario generation considering decision error is proposed. • An interval mean-CVaR model is established. • The impact of interval uncertainty on offering strategy and income is analyzed.