Estimation of Battery Parameters in Cascaded Half-Bridge Converters with Reduced Voltage Sensors

Although modular multilevel converters (MMC) and cascaded half-bridge (CHB) converters are an established concept in HVDC, MMCs and CHBs have started to find new applications, including modular converters with integrated energy storage systems. Despite various advantages of the low-voltage modularity, a complex and expensive monitoring/control system can hinder finding a foothold in many emerging applications that are more cost-driven, such as the e-mobility market. Estimators and observers can reduce the monitoring cost and complexity by reducing the number of required sensors and communication bandwidth. However, estimation methods rarely consider MMCs with integrated battery, and most available methods neglect all resistances. This paper fills this gap by developing an online estimation technique for parameters of all battery modules in an MMC. The proposed method exploits the slow dynamics of the battery to use a simpler and less computationally demanding algorithm that can easily be implemented in low-end controllers. Based on the developed model of the system, the iterative algorithm can estimate the voltage and internal resistance of every module through measuring the output voltage and current of the battery pack and avoid direct measurements from the modules. As a result of substantial reduction in the number of monitoring sensors for estimating the battery parameters, the proposed technique is simpler and less costly in comparison with other sensor-based techniques. Furthermore, the proposed technique accelerates convergence using optimal learning rate value. Simulations validate the ability of the proposed estimation technique under different scenarios. The estimation technique can identify both internal resistance and open-circuit voltage of the batteries with approximately 2 % accuracy.