Faciliating the Service of a Shared PV and Energy Storage System Considering Users’ Discrete Choice

Even though the shared energy resource (SER) is in service in the community, the description of the stochastic and discrete nature of user choices is commonly omitted in the formulation of the SER, which may trigger uneconomic supply-related decisions to be made by its owner. The discrete choice model (DCM) is capable of describing users’ uncertain preferences. However, the disaggregate demand representation derived from DCM is nonlinear and nonconvex, making it challenging to embed into the MILP problem. In this paper, we overcome the probabilistic nature of user DCM by leveraging sample average approximation (SAA) to specify the demand directly in terms of the utility functions instead of the choice probabilities. In addition, we propose a bi-level framework based on the Stackelberg game that can explicitly capture the interplay between the behavior of consumers and the decisions of the SER owner to be made. Numerical experiments are further deployed to perform sensitivity analysis regarding sample size, user type, and shared energy storage capacity. The results indicate that the proposed method is able to model user preferences and help the owner to develop a reasonable pricing and operation strategy.