Spreads

General Description

Each cell in a battery has a cell voltage that correlates with that individual cell's State of Charge (SOC). Since no two cells are exactly the same due to manufacturing tolerances, aging, and environmental factors, a (dis)charge signal applied to a battery module or rack affects each cell slightly differently. Over time, these differences accumulate, leading to variations in cell voltage, known as cell imbalances.

If the difference in cell voltages becomes too large, it not only reduces the usable capacity of the battery system—because some cells will reach their voltage limits earlier than others—but it can also compromise the performance, safety, and overall lifespan of the battery. This can eventually make the system harder to control, increase maintenance costs, and lead to premature failure or downtime. The same problem can occur for temperature and State of Charge.

Spreads are useful for analyzing individual components rather than comparing performance across different component levels.

Voltage and SoC Spreads

Conceptionally, the cell voltage spread is the voltage difference between the cell with the maximum voltage and the cell with the minimum voltage. However, the voltage spreads are normalized to a certain voltage range and conditioned to other requirements. This means, voltage spreads are not just simply maximum minus minimum voltage. The calculation is only done, if certain requirements are fulfilled, these are as default settings i.e.:

• relaxed voltage is between 3.20 and 3.25 V for LFP or 3.50 and 3.75 V for NMC

• relaxation time is above 120 sec

• current is below a certain threshold of a couple of Amps

The SoC spread is the SoC difference between the cell with the maximum SoC and the cell with the minimum SoC. Considering that we are using the TWAICE SoC for the SoC spread, we have no requirement on the operation, only that TWAICE Cell SoC min and TWAICE Cell SoC max need to be available.

Temperature Spreads

The module temperature spread is the temperature difference between the cell with the maximum temperature and the cell with the minimum temperature.

TWAICE Spreads Calculation

How does TWAICE calculate spreads?

Data is automatically preprocessed using the TWAICE preprocessing pipeline. Once the data is cleaned, the spread value is calculated for each timestamp and stored. Finally, the maximum of all spread values is returned, i.e. to the TWAICE UI.

Advantages of TWAICE spread calculation

Spreads can indicate issues with the battery quickly as they are state-independent and need no convergence. However, only relying on raw data will lead to faulty indications due to data quality issues. These issues are solved by using TWAICE data preprocessing and only reliable data points are used. For example, given that the voltage is SoC dependent, voltage spreads are only meaningful if the data is properly preprocessed and put into perspective and compared in a correct way.

This means that problems within the battery can be seen instantaneously.

What you can do with the TWAICE spreads?

Evaluation of cooling / HVAC system

Various issues manifest themselves via inhomogeneous temperature distributions, for example insufficient thermal management for the applied operation. This is particularly relevant if specific high power patterns are applied, as these regularly lead to large temperature spreads and stress on the batteries.

Another example could be a poor thermal management design which can be identified by correlating temperature spreads with locations and component structure which happen continuously throughout the operation.

Along with the examples mentioned above, defect cooling system elements such as fan outages can be identified due to high temperature spreads and overall high temperatures, which might be dependent on the exact location of the temperature spread.

Identification of weak cells and self-discharge

Weak cells refers to cells which exhibit a lower performance compared to other cells. These cells show usually a higher DCR resistance and (which can be identified using the TWAICE Live DCR if cell level data is available) voltage, SoC or temperature deviations to the other cells because it exhibits lower efficiency. Having these different KPIs enables the user to quickly identify specific cells.

Regarding self-discharge, specific internal processes, manufacturing issues or aging mechanisms can lead to a higher self-discharge than what is usually observed. These cells develop a increasing voltage and SoC deviations. So especially having a view over time facilitates the identification.

Detection of balancing issues and efficiency

In the planned operation mode, the Battery Management System (BMS) or Energy Management System (EMS), is balancing the battery components to ensure they exhibit the same temperature, voltage and SoC. However, these systems regularly show malfunctions. Therefore, having a view on these KPIs ensures a check on the BMS and EMS functionality.

In addition, these KPIs provide an evaluation on the efficiency once a balancing has been performed. Ideally, after a balancing, the entire system shows homogeneous distribution of these KPIs. Leveraging the discussed spread KPIs, the efficiency of such a balancing process can be evaluated.