AI-based site selection and operational optimization of large-scale battery storage systems to ensure grid stability and potentially reduce the need for grid expansion (smartBattery)
The smartBattery project is developing an AI-based method for optimized site selection and operational management of large-scale battery storage systems to promote grid stability and integration of renewable energies and thus potentially reduce the need for grid expansion.
The project addresses both the challenges faced by electricity grid operators and operators of large-scale battery storage systems. Grid operators, due to grid limitations and the resulting need for grid expansion, cannot implement the large number of grid connection requests for large-scale battery storage systems in a timely manner. Since the operation of large-scale battery storage systems is only market-oriented and possibly system-serving, high safety margins must be expected when applying for grid connection. Storage project developers and operators, on the other hand, have a strong interest in connecting their systems to the grid as quickly and cost-effectively as possible. Their challenge also lies in identifying suitable locations and grid connection capacities.
Objective
The aim of the project is to resolve this discrepancy. Through the active designation of grid-serving or grid-neutral locations by the grid operator, planning security can be offered to both sides. The consideration of operational constraints enables flexible large-scale battery storage systems, in addition to marketing on existing markets (in particular the spot market and balancing energy), to also take the local grid status into account.
Procedure
The intended solution approach lies in a stakeholder-neutral, transparent evaluation methodology that reflects all interests involved. In the smartBattery Toolbox, data-driven methods simulate, optimize and assess various operating scenarios from the perspective of grid operators and storage operators based on the provided input data and boundary conditions. Through the use of machine learning and optimization models, suitable locations are determined. In addition, an optimized storage dimensioning and the definition of operational constraints are carried out.
By simultaneously taking into account grid restrictions and revenue potentials, a significant contribution can be made to the integration of renewable energies and the efficient use of grid infrastructure. This innovative approach thus offers relevant added value for storage operators and grid operators, but also for the economy, ecology, security of supply and climate protection.
