Interpretación de los resultados obtenidos de las encuestas realizadas a los clientes

Inaccurate OCV and OCV hysteresis data will lead to an inaccuracy in SoC estimation. An inaccurate SoC will be reflected as inaccurate range estimation, leading to decrease of user satisfaction/trust; which in turn is a potential business risk to the OEMs. On the other hand, the inaccuracy in SoC can lead to shift of operating SoC window of EV’s battery packs. To maintain minimum available power assist and regenerative capability, HEV battery packs operate within a SoC window, avoiding high and low SoC [136]. A SoC window is also used for other types of EVs to extend battery life and avoid safety failures due to overcharge and over-discharge [136, 137]. An inaccurate measurement of SoC can shift/change the operating SoC window which will be reflected as short term (e.g. regenerative power capability) and long term performance drop (e.g. decrease of expected battery life). Therefore, it is important that the ECM used by BMS should incorporate an accurate OCV.

In the step OCV test procedure, OCV is measured while the battery is incrementally charged (ܱܥܸ_ୡ) from fully discharged state or incrementally discharged(ܱܥܸ_ୢ) from fully charged state. The OCVs can then be associated with the corresponding SoC. SoC is calculated from the charge (ܳ_௖) or discharge (ܳ_௖) capacity that has been added to or removed from the cell. The capacities are defined as:

ܳ௖(ݐ) =න ܫ௖(ݐ)݀ݐ ௧ ଴ (16) ܳௗ(ݐ) =න ܫௗ(ݐ)݀ݐ ௧ ଴ (17)

In Equations (16) and (17) ܫ_௖ and ܫ_ௗare charge and discharge currents and are assumed positive in value. Battery SoC is assumed to be 0 % when the cell voltage (not OCV) reaches the minimum discharge voltage and 100 % at the end of CV part of charging. Depending on the charge/discharge conditions e.g. rate, temperature, intermediate rest period, age etc.ܳ_௖/ܳ_ௗ can be different. From the origin of OCV (as explained in Section 2.3) it is expected that OCV will be directly related to the capacity not the empirically determined SoC. This will be investigated further and reported in Chapter 8.

To investigate OCV hysteresis,ܱܥܸ_ୡandܱܥܸ_ୢ are required to be compared against a common capacity axis, instead of two separate capacity axesܳ_௖andܳ_ௗ. Therefore, an initial condition can be introduced and the current ܫcan be assumed positive for discharge and negative for charge. The common capacity scale can be identified as the remaining capacity (ܳ_௥), is now defined as:

ܳ௥(ݐ) =ܳ௥(0)−න ܫ(ݐ)݀ݐ ௧ ଴ ܫ(ݐ) > 0 Discharge

ܫ(ݐ) < 0 Charge

When ܱܥܸ_ୡand OCV_ୢcurves are plotted against the common axis,ܳ_௥, an erroneous hysteretic behaviour may be observed. The apparent hysteresis artefact arises due to the testing procedure and in the assumption that the remaining capacity is zero (ܳ_௥(0) = 0) at the end of the discharge prior to the start of the ܱܥܸ_ୡ test. For example, the cell needs to be discharged prior to the ܱܥܸ_ୡ characterisation test, for which a 1C constant current discharge can be performed up to the cell cut-off voltageܸ_{௠ ௜௡}. The test is terminated and first OCV is measured as the starting value of the ܱܥܸ_ୡ test and the remaining cell capacity (ܳ_௥) is assumed zero. In comparison, during the ܱܥܸ_ୢ test, diffusion limitations are reduced as the cell is discharged incrementally toܸ_{௠ ௜௡}, and this allows for more capacity to be removed before the cell reachesܸ_{௠ ௜௡}. Later in this thesis (Chapter 8) will be demonstrated the variation of discharge capacity with different step sizes. The remaining cell capacity (ܳ௥)will then again be assumed zero (since the cell reachedܸ௠ ௜௡for theܱܥܸୢtest)

however, the measured OCV value after similar rest will be lower (due to more capacity removal) in comparison to the starting value of the ܱܥܸ_ୡ test. Thus, when plottingܱܥܸ_ୡandܱܥܸ_ୢagainst remaining capacity an offset between the curves will be present, invalidating any true hysteresis assessment. Also, this phenomenon will have significant effect on repeatability and reproducibility of the ܱܥܸ_ୡ and ܱܥܸ_ୢ curves. However, researchers reporting OCV and OCV hysteresis [114, 115, 121- 125, 127, 138], did not consider this phenomenon. Therefore, a robust methodology to assess OCV and OCV hysteresis is missing; consequently, an erroneous assessment of OCV and OCV hysteresis could be present historically.

In addition to that, active material particles with a non-monotonic chemical potential (as explained in Section 3.4.1) are expected in many intercalation battery systems and not only restricted to lithium iron phosphate electrodes. As such, a certain magnitude of hysteresis could also be present in other insertion electrochemical systems, which is yet unknown in the scientific community. As explained in Chapters 1 and 2 not only LFP, but also other Li-ion battery chemistries also considered for automotive applications, therefore it is important to fill this gap in knowledge.

Automotive application requires a wide range of operating temperature and theoretically battery OCV is dependent on temperature (Section 2.3 and Equation (6)). However, Pattipati et al. concluded no change of OCV-SoC relationship with temperature [117]. Also, higher level of OCV hysteresis with lower current rate has been reported previously by Roscher et al. [124]. Therefore, it is important to investigate change of OCV and hysteresis with temperature and C rate to incorporate accurate information to BMS.

Although a standard OCV test is missing in existing standards, researchers previously used different approaches to measure OCV. Step OCV test is appropriate for accurate measurement, however, following issues required to be addressed:

1. Effect of step size on OCV measurement needs to be investigated 2. Use of OCV vs capacity not SoC needs further research attention 3. OCV change with temperature requires to be identified

4. OCV change with discharge rate needs to be investigated 5. OCV hysteresis of different chemistry cells needs to be studied

4.4 Summary

The existing standard capacity tests with constant current discharge at C rates and temperatures do not reflect the dynamic discharge profile of an automotive battery pack and current/power de-rating. Hence, constant current discharge capacity requires to be validated with dynamic discharge profile e.g. drive cycle for automotive application. Also, commonly a BMS calculates a battery’s SoC through coulomb counting and SoC is used to estimate RDR. Although different approaches have been proposed in literature to translate SoC to RDR, discrepancy of RDR estimation in existing EVs is still require significant improvement. This is possibly due to that fact that SoE, not SoC, has a direct relationship with RDR. This thesis will focus on use of SoE in RDR estimation.

Since EIS is considered as an online measurement technique, ECM parameterisation method and periodic characterisation tool as part of long duration ageing tests, a reliable and reproducible measurement is essential. Inadequate relaxation period can potentially significantly affect the reproducibility of the test result. This thesis will provide insight to the effect of short and long term relaxation period on EIS results at different temperature and SoC.

To develop an ideal OCV test protocol, the influence of step size on OCV measurements will be researched as part of this thesis. The proposed testing protocol then will be used to identify OCV and thus research OCV hysteresis of different chemistry lithium-ion cells, including different temperatures and charge-discharge rates.