Click on each author name to view the details.
PV plant reliability depends on energy storage reliability. Despite all the new storage technologies, lead-acid batteries remain the most used in off-grid PV systems because they have the best cost-benefit ratio. For remote PV plants, due to the time length process of sending spare parts besides the maintenance cost, it is essential having a way to predict the end-of-life (EoL) of batteries. There is a need for a non-invasive, simple and robust method. Non-invasive means that an algorithm must be capable of testing the battery without interrupting its continuous operation. It must be simple, minimizing complex devices or arrangements inside PV plant, and it must be robust to work in rough environment of remote field, prone to electrical noise and working without human presence during months. Another concern is cost for simultaneous monitoring of several batteries in PV plant. For predicting battery’s EoL, there are several algorithms. However, most of them are suited for laboratory use and others for vehicular use. The way they could be used for remote PV plants and for monitoring lots of batteries is not clear. The common approach for battery parameters estimation, e.g. State-of-Health (SoH) is based on equivalent circuit models that are mathematically represented by state space equations. Unlike lead-acid batteries used in backup systems that operate most of the time with 100% of State-of-Charge) (SoC), the SoC and SoH estimation in off-grid PV systems is a hard task, in reason of the uneven cycling regime and the rare full charge of the batteries.