Forecast-based charging strategy to prolong the lifetime of lithium-ion batteries in SHS


Standalone PV battery systems have great potential to power the one billion people worldwide who lack access to electricity. Due to remoteness and poverty, durable and inexpensive systems are required for a broad range of applications. However, today’s PV battery systems do not yet fully meet this requirement. Especially batteries still prove to be a hindrance, as they represent the most expensive and fastestaging component in a PV battery system.

This study aims to address this by prolonging battery life. For this purpose, a forecast-based charging strategy was developed. As lithium-ion batteries age slower in a low state of charge, the goal of the operation strategy is to only charge the battery as much as needed. The impact of the proposed charging strategy is examined in a case study using one year of historical data of 14 standalone systems in Nigeria.

It was found that the proposed operation strategy could reduce the average battery state of charge by around 20% without causing power outages for the mini-grids. This would significantly extend the life of the battery and ultimately lead to a more durable and cheaper operation of standalone PV battery systems

As the aging of lithium-ion batteries is accelerated when they are operated in a high state of charges, the proposed strategy pursues the goal of only charging the battery as much as necessary. To mitigate the risk of a dead battery and potential power outages, various safeguards are added to the algorithm.

The reduction of the SOC is accomplished by two approaches. On one hand, the charge of the battery is postponed from the morning to the afternoon. On the other hand, an upper SOC cap, above which the battery is stopped being charged, is introduced. With the help of PV generation and consumption forecasts, the algorithm decides to which extent each of the approaches shall be used. As forecasts are not always accurate, an adjustable buffer is used as a safeguard.

To evaluate the impact of the proposed strategy, a case study with 14 standalone PV battery systems powering a local market in Nigeria was conducted. Simulating the operation of considered systems for one year showed that the proposed charging strategy can be beneficial. Without causing additional power outages, the new strategy reduces the average SOC by ≈ 20% for most systems while not causing additional power outages. Similarly, the average time per day in which the battery is fully charged was reduced by ≈ 7 hours. When tuning the strategy more dynamic, the SOC can be reduced even further. However, this comes at the cost that the battery is more often fully discharged which leads to power outages. Further research should investigate the period by which lowering the SOC through this operating strategy extends the life of an LFP battery.

The results of this work can be extended to any standalone PV battery system and are not geographically limited. Systems with high average SOCs and low utilization of the battery can profit most, as the average SOC can be reduced to a greater extent, which in turn leads to a higher battery lifetime. Finally, this new forecast-based operation strategy may be of particular interest for batteries in tropical climates, as fast battery degradation poses a greater problem there.”


Download the full study.