A high-performance lithium-ion batteries state of charge estimation strategy for photovoltaics application

Abstract

Abstract Virtual power plants (VPPs) can realize the aggregation and coordinated optimization of distributed generation (DG), such as renewable energy (e.g. photovoltaic (PV), wind power) and energy storage systems (ESSs), which is considered an up-and-coming technology. The lithium-ion battery is widely used in VPPs as a high-quality energy storage. Meanwhile, the battery state of charge (SOC) estimation is fundamental, which characterizes the remaining energy, and obtaining an accurate battery SOC is essential for the safe and reliable operation of the ESS. This article proposes a SOC estimation technique for lithium-ion batteries in VPPs containing PV. Considering that the extended Kalman filtering (EKF) can obtain a more accurate battery SOC, it is further optimized based on the EKF to improve the accuracy of its SOC estimation. First, the principle of the general method and its deficiencies are analyzed, and then the proposed improved method is analyzed in detail. Finally, the advantages of the proposed method and the feasibility of applying it to PV-ESSs are analyzed in detail. A test platform is built for validation, and the results show that the SOC estimation errors of the proposed strategy are 1.8%, 2.1%, and 2.3%, respectively, with better performance.