State-of-health estimation of batteries in an energy storage system based on the actual operating parameters

Abstract The battery state-of-health (SOH) in a 20 kW/100 kW h energy storage system consisting of retired bus batteries is estimated based on charging voltage data in constant power operation processes. The operation mode of peak shaving and valley filling in the energy storage system is described in detail. Two SOH modeling methods including incremental capacity analysis (ICA) and probability density function (PDF) are compared. The results show that such battery SOH modeling methods as ICA and PDF are available under constant power conditions. The SOH estimation model by PDF method has an enhanced accuracy on the basis of the same voltage range. The height H of the largest peak in ICA or PDF curves has a linear positive correlation with the battery SOH, but the SOH model by ICA is more accurate than that by PDF. According to the SOH evaluation, the energy storage of the system will be significantly improved if some cells or modules with lower SOH are replaced by those with higher SOH. In the condition of unknown SOH of battery, the relative aging degree of battery can be estimated by grading the H value from ICA or PDF curves based on actual charging voltage data, which is helpful for the operation and maintenance of the energy storage system.