A Stochastic Multi-Temporal Optimal Power Flow Approach for the Management of Grid Connected Storage

Renewable energy (RE) integration into distribution grids is becoming more common in the context of the energy transition. The management of wind or solar generation due to their variability and low predictability are challenging for distribution system operators (DSO). To that may be added uncertainties related to electric load profiles. The role of flexibility, coming from decentralized storage devices, will be important for DSOs trying to manage uncertain loads as well as high levels of RE penetration. The introduction of automation and smart metering in distribution grids allows for the optimized management of storage devices to maximize the capability of current infrastructure to integrate RE generators. These optimized management strategies can be calculated with optimal power flow (OPF) algorithms. This paper uses a convex relaxation of the power flow equations to expand the multi-temporal deterministic approach presented in [1] to a stochastic one. The stochastic algorithm implies the integration of a scenario tree to plan the charging and discharging schedule of batteries one day in advance. When comparing deterministic and stochastic operation planning strategies, the stochastic method annually increases total economic benefit by 3.1% while requiring lower annual cycling of the battery therefore increasing battery life.