Practical state estimation using Kalman filter methods for large-scale battery systems

Abstract

The system states of battery energy storage systems (BESSs) such as state of charge (SOC) and state of health (SOH) are essential for the functions of the system, such as frequency support services and energy trading. However, the complexity of a large-scale battery system makes the estimations more difficult than at the cell-level. This is further compounded by real-world limitations on system monitoring data granularity, accuracy and quality. In this paper it is shown how cell-level state estimation techniques can be utilised on large-scale BESSs using experimental data from a 2MW, 1MWh BESS. The results show how a Dual Sigma point Kalman Filter (DSPKF) SOC estimation provides more accurate results compared to the commercial BESS battery management system SOC. It is shown how the DSPKF parameters can be tuned by a genetic algorithm to simplify selection and generalise the approach for different BESSs. Furthermore, it shows how this method of SOC estimation can be combined with a total least-squares (TLS) method for capacity estimation to less than 1% error. Online system state estimation is demonstrated using both designed tests and real-world operational profiles where the BESS has provided contracted frequency response services to the national electricity grid in the UK.