Why SOH decreases in few cycles?

Good Morning,

I used this model to estimate the state of health but I noticed that the battery reaches its end of life rapidly, in few cycles of charging and discharging. Someone know why? Because in my opinion it is not realistic. Which parameters could I modify in order to perform a more realistic analysis? Thank You.

Answer 1

[Xiangchun]

Hi Irene,

Could you please provide more details on the model referenced?

It is true that Li-ion batteries in the real world do not reach end of life in just a few cycles. In general, the examples provided in the documentation specify aggressive battery degradation parameters so that the aging effect does not take hundreds of cycles to be noticeable. These examples are for illustrative purpose, and the aggressive degradation parameters reduce the required simulation time.

The sections of "Fade" and "Calendar Aging" provides detailed explanation on how to set up model parameters for cycling and calendar aging in the Battery (Table-Based) block. Tabulated battery model - MATLAB (mathworks.com)

Best wishes,

Xiangchun

[irene giusti]

Hi Xiangchun,

Thanks for your answer, I disabled the calendar aging and for what concerns the fade I set 'equation' and number of discharge cycles: 100 (instead of 10).

I have set charging and discharging cycles at 1C between 25% and 75% SOC; in this way I reach the 80% of SOH in about 80 cycles, which is low with respect to the 1000 cycles expected for a battery used in EV (I am analysing a 4680 cell type).

Which parameters need to be changed in order to reach more realistic results? Does the problem come from the setting of R0_mat, R1_mat and tau1_mat (I did not change them)?

Thank you in advance.

Kind Regards,

Irene

[Xiangchun]

Hi Irene,

It seems that this is the example you are referring to: Battery State-of-Health Estimation - MATLAB & Simulink (mathworks.com).

There are two things relevant to cycle life: (1) how fade (cycling aging) is set up in the Battery (Table-Based) block, and (2) how SOH is calculated from the "SOH Estimator" block.

In the Fade parameterization, if "Equation" based method is used, in addition to the "Number of discharge cycles," "Change in terminal resistance after N discharge cycles (%)" can be adjusted to reflect the cycling aging behavior of the 4680 batteries you are working with. Reduce this percentage would slow down the aging progress from the terminal resistance standpoint. This dictates how fast terminal resistance changes cycle on cycle according the following equation from the documentation page Tabulated battery model - MATLAB (mathworks.com).

From the "SOH Estimation" block, it is implemented in such a way that SOH = 0% when terminal resistance reached "Terminal resistance at end of life." In the shipping example Battery State-of-Health Estimation - MATLAB & Simulink (mathworks.com), terminal resistance at end of life is specified as "R0_mat*3." This factor of 3 can be adjusted (increased) to reflect the batteries you are working with and your preferred definition of end of life. This dictates how SOH at end of life is defined.

Best wishes,

Xiangchun

[irene giusti]

Thank you for the answer, now it is clear how to proceed.

Irene