Data-Driven Model Predictive Control Strategy for Battery Energy Storage System

The battery energy storage system (BESS) is a large-scale battery system used for storing electricity and energy. This paper proposed a data-driven method to model the battery system and uses the Koopman operator to obtain a linearized model of the nonlinear BESS. In the scheme, deep learning methods are used to identify Koopman operators and obtain an approximated finite-dimensional Koopman space at the same time. The Koopman linearized system is controlled based on the model predictive control (MPC) method. Simulation results demonstrate the effectiveness of the proposed data-driven Koopman MPC strategy.