Virtual Calibration of Fast Charging Strategies in GT-AutoLion

Written by Joe Wimmer

June 29, 2020
Electric Bus

As mentioned in a previous blog, charging strategies have a significant impact on battery longevity and market perception, which heavily effects the market’s overall perception of a brand.

Depending on the complexity of a charging strategy, it may require optimization, or calibration.  is extremely difficult, expensive, and time consuming.  Intuitively, this is true because of the time and cost associated with physical testing.  Less intuitively, however, this is also true because experimental testing of Li-ion cells can only measure voltage, current, and temperature of the cell or battery; whereas, understanding the true impact of a charging strategy requires much more detailed information about the state of the battery.

Luckily, physics-based modeling of Li-ion cells using GT-AutoLion enables engineers to have insight into the electrochemistry inside Li-ion cells well beyond voltage, current, and temperature measurements. This ultimately enables charging strategies to be virtually calibrated.

GT-AutoLion not only calculates the high-level quantities of Li-ion cells like voltage, current, and temperature, but also hundreds of other physical quantities within the cell, allowing engineers to have valuable insights into the electrochemistry of the cell.  With this ability, GT-AutoLion enables engineers to explore, calibrate, and make robust decisions for fast charging algorithms early in a development cycle, enabling less physical testing to be done and when physical testing needs to be done, the tests are better informed and more focused.

Example Quantities

Thanks to a discretization of the anode, separator, and cathode in a one-dimensional manner, GT-AutoLion solves for many quantities throughout the thickness of the cell.  For instance, the electrical potential of the electrolyte and the concentration of Lithium-ions throughout the thickness of the cell are solved for on a location-dependent and time-dependent manner, as shown in the image below, which summarize the results of an example 1C discharge event (Left) and a 1C charge event (Right).  These quantities help give insight into electrochemical performance within the cell. The potential in the electrolyte at various times is represented in the top plots and the concentration of Lithium-ions at various times is represented in the bottom plots.

1C Discharge (Left) and Charge (Right) Results – The space to the left of the first dashed vertical line represents the anode, the space between the two dashed vertical lines represents the separator, and the space to the right of the dashed vertical lines represents the cathode

Electrolyte Potential & Fast Charging Strategies

As mentioned in a previous blog, charging Li-ion cells too fast can lead to premature degradation of the cell primarily due to lithium plating in the anode.  Lithium plating occurs in the anode when the electrolyte potential is above zero in that electrode.  This is illustrated in the image below which plots the electrolyte potential across the anode, separator, and cathode at various times.

Lithium Plating Chart
Plot showing electrolyte potential during a 1C charge event and the area that would cause lithium plating to occur

This simulation and plot can be repeated for higher charging rates, including 1C (charging the cell in one hour), 2C (30 minutes), and 3C (20 minutes).  The results are summarized in the image below.  Clearly, there is a high risk of lithium plating when charging this particular Li-ion cell at a constant current of 3C.

Electrolyte Potential plot for 3 different charge rates: 1C, 2C, and 3C

Virtual Sensors

These quantities calculated by GT-AutoLion are not only available to be plotted after a simulation, but can also be dynamically sensed and even used in controls algorithms.  GT-AutoLion provides an easy-to-use framework for sensing any quantity at any location, which can then be sent to GT’s controls domain or even 3rd party tools specialized in controls for use in controls development (for Model in the Loop or MiL testing).  With this framework, more complex charging strategies can be explored and calibrated.


Virtual calibration of batteries enables real-time or faster than real-time virtual battery sensors or observers that can be used to optimize charging strategies.  Additionally, this framework enables engineers to understand the root cause for battery degradation and even anticipate early failures.

The primary and most immediate business benefit early in the development cycle is a reduction in testing costs by simply minimizing cell aging tests.  This means product development will be faster and more cost-effective.

From a marketing and product desirability point of view, performance metrics can be met and potential critical lifetime scenarios averted. This, over time, translates to improved brand perception and greater long-term brand loyalty.

Written By: Joe Wimmer