A photovoltaic–storage system configuration and operation optimization model based on model predictive control

Abstract The deployment of distributed photovoltaic technology is of paramount importance for developing a novel power system architecture wherein renewable energy constitutes the primary energy source. This paper investigates the construction and operation of a residential photovoltaic energy storage system in the context of the current step–peak–valley tariff system. Firstly, an introduction to the structure of the photovoltaic–energy storage system and the associated tariff system will be provided. Secondly, to minimize the investment and annual operational and maintenance costs of the photovoltaic–energy storage system, an optimal capacity allocation model for photovoltaic and storage is established, which serves as the foundation for the two-layer operation optimization model. And the installed capacity of photovoltaic and energy storage is derived from the capacity allocation model and utilized as the fundamental parameter in the operation optimization model. Furthermore, taking into account the impact of the step–peak–valley tariff on the user’s long-term energy use strategy, a two-layer optimization operation algorithm for the photovoltaic–storage system based on model predictive control is proposed. The upper model is an annual optimization based on the step tariff, to maximize the annual comprehensive revenue. The lower model is a daily rolling optimization based on the peak–valley tariff, to minimize the daily operation cost. The operation schemes of the photovoltaic system and energy storage in the lower layer model utilize the upper layer optimization results as a reference point, correcting for any deviations in the system state due to uncertainty factors. Ultimately, the results of the arithmetic simulation demonstrate that the proposed models can delay the introduction of high-step tariffs and significantly enhance the overall benefit to residential users.