CONTINUOUS LOAD STATE ESTIMATION DEVICE INTEGRATING A CORRECTIVE FACTOR, VEHICLE, METHOD, AND PROGRAM BASED ON SUCH A DEVICE

Abstract

The invention relates to a device for estimating the state of charge of a battery connected to the device, the battery comprising battery cells having a positive electrode comprising lithium iron phosphate. The estimation device comprises: - a correction means (M1) determining a correction factor based on the difference between a precise state of charge estimated on the basis of the open-circuit voltage, and the corresponding approximate state of charge estimated as a function of a previous approximate state of charge, the cell current, and the cell capacity; - an estimation means (M2) estimating the current state of charge as a function of a previous approximate state of charge, the cell current, and the cell capacity, including said correction factor. The invention also relates to a vehicle, a method, and a program based on such a device. Figure 2

Description

Title of the invention: CONTINUOUS CHARGE STATE ESTIMATION DEVICE INTEGRATING A FACTOR CORRECTIVE, VEHICLE, METHOD, AND PROGRAM BASED ON SUCH A DEVICE

[0001] The invention relates to the field of devices and methods for estimating the state of charge of a battery, in particular a traction battery of a motor vehicle.

[0002] The traction battery consists of Li-Ion cells. These cells have an authorized operating range within which they must remain to ensure their proper operation over time, and the absence of safety events such as a fire.

[0003] An essential quantity is the state of charge. This quantity represents the level of electrical charge of each cell.

[0004] Generally, the higher the state of charge, the more the charging current must be limited. It is therefore important to accurately estimate the state of charge. If the state of charge is underestimated, current levels that can damage the battery will be permitted. The inaccuracy of the estimator is then taken into account. This naturally has an impact on performance in terms of current level and therefore charging time: the greater the uncertainty, the greater the impact on performance.

[0005] Conversely, if the state of charge is overestimated, this may lead the user to break down due to lack of charge in an area without a charging point (highway, rural area etc.).

[0006] Fig. 1 represents the open-circuit voltage OCV of the cells on the ordinate and abscissa, the state of charge SOC of the battery cells having NMC technology cells (for nickel-manganese-cobalt positive electrode) in solid line; and LFP technology cells (for lithium iron phosphate positive electrode) in dashed line.

[0007] The state of charge is estimated in several ways: - it is regularly recalibrated based on the open-circuit voltage of each cell after relaxation. This is the most accurate measurement. It is also referred to as open-circuit voltage, or "OCV" for "open circuit voltage" in English; - between these recalibration opportunities, while the vehicle is in motion: a) The state of charge is mainly estimated by integrating the incoming and outgoing current using the Coulomb balance method. This integration technique is flawed by an error that grows over time because the method integrates, that is to say adds, errors continuously, particularly on the fine measurement of the current; b) and if the open-circuit voltage curve of the NMC cells has a sufficient slope ([Fig.1] defined below), then the battery management system can also, in addition to the coulomb balance, correct the state of charge on the basis of the open-circuit voltage and a Kalman filter, but this remains less precise than the regular recalibration of the state of charge on the basis of the open-circuit voltage.

[0008] The drop in open-circuit voltage is flatter for LFP cells than for NMC cells, which has an effect on the estimation of the state of charge.

[0009] The invention relates to the configuration of the LFP technology. Recalibration of the state of charge by the open-circuit voltage is not possible in a significant region of the state of charge, because the curve is flat.

[0010] The only real opportunities to learn the state of charge from the open-circuit voltage (OCV) measurements of the cells are represented by circles in [Fig. 1]. These are the areas where the open-circuit voltage varies significantly with the voltage.

[0011] The SOC > 95% zone is particularly noteworthy. Therefore, it is common practice to recommend that users of vehicles with LFP battery technology regularly charge their vehicles to 100%. If they do not, their uncertainty regarding the state of charge will increase with use.

[0012] However, this uncertainty of state of charge must be taken into account in the determination of the limits of the cells related to the state of charge, so as never to underestimate the state of charge.

[0013] When the entire system may exhibit high state-of-charge uncertainty, this must be taken into account; that is, the uncertainty must be considered and margins adjusted accordingly. This naturally has significant impacts on performance, for example, in terms of charging time and the possibility of recovering braking energy.

[0014] An objective of the present invention is to remedy the defects of the prior art, and in particular to offer a solution for estimating the state of charge of a battery that is more accurate than in the prior art.

[0015] To achieve this objective, the invention proposes a device for estimating the state of charge of a battery connected to the device, the battery comprising battery cells having a positive electrode comprising lithium iron phosphate, the estimation device comprising: - a correction means determining a correction factor based on the difference between a precise state of charge estimated on the basis of the open-circuit voltage, and the corresponding approximate state of charge estimated as a function of a previous approximate state of charge, the cell current and the cell capacity; - an estimation method estimating the current state of charge based on a previous approximate state of charge, the cell current and the cell capacity, in including said correction factor.

[0016] Advantageously, the invention reduces the uncertainty of the state of charge and offers maximum performance in terms of autonomy and power to all users. It also compensates for the potential effects of aging.

[0017] According to one variant, the estimation means estimates the current state of charge based on the formula: SOC(t) = SOC(tA f) - °Ù SOC(t) is the current state of charge; SOC(t-St) is the preceding approximate state of charge; A t is the difference between a current time and a previous time; a is the correction factor; I is the current in amperes; This is the capacity in As

[0018] This allows for an accurate estimation of the current state of charge.

[0019] According to one variant, the correction factor is a current offset error between the precise state of charge estimated on the basis of the open-circuit voltage, and the corresponding approximate state of charge.

[0020] This makes it easy to determine the correction factor.

[0021] According to one variant, the correction factor is half of a current offset error between the precise state of charge estimated on the basis of the open-circuit voltage, and the corresponding approximate state of charge.

[0022] This improves the accuracy of the correction factor.

[0023] According to one variant, the correction factor is determined using the formula a = % = * A a is the correction factor; d is the offset error; A t is the difference between a current time and a previous time; A SOC is the difference in state of charge between the current time and a previous time.

[0024] This allows for an accurate estimation of the correction factor.

[0025] The invention further relates to a battery management system comprising an estimation device according to the invention.

[0026] Another object of the invention relates to a motor vehicle comprising: - a traction battery comprising battery cells having a positive electrode comprising lithium iron phosphate; and - an estimation device according to the invention, connected to said traction battery.

[0027] The invention also relates to a method for estimating the state of charge of a battery comprising battery cells having a positive electrode comprising lithium iron phosphate, by means of an estimation device according to the invention, the estimation method comprising the following steps: - a correction step in which a correction factor is determined based on the difference between the precise state of charge estimated on the basis of the open-circuit voltage, and the corresponding approximate state of charge estimated as a function of a previous approximate state of charge, the cell current and the cell capacity; - an estimation step in which the current state of charge is estimated as a function of a previous approximate state of charge, the cell current and the cell capacity, including said correction factor.

[0028] According to one variant, in the estimation step, the current state of charge is estimated on the basis of the formula: SOCSOC(t-to t)- where SOC(t) is the current state of charge; SOC ( LA / ) is the preceding approximate state of charge; A t is the difference between a current time and a previous time; a is the correction factor; I is the current in amperes; This is the capacity in As

[0029] The invention further relates to a computer program comprising program code instructions for executing the steps of the estimation process according to the invention, when said program is running on a computer.

[0030] The invention will be further detailed by describing non-limiting embodiments, and based on the accompanying figures illustrating variants of the invention, in which: - [Fig. 1] schematically illustrates a graph of the open-circuit voltage and state of charge of battery cells using NMC (solid line) and LFP (dashed line) technologies; and - [Fig.2] schematically illustrates the device and estimation method according to the invention.

[0031] The invention relates to a device for estimating the state of charge of a battery, in particular a traction battery of a motor vehicle.

[0032] The invention consists, during no-load voltage measurements, of learning the errors that cause the uncertainty in the estimation of the state of charge by integration and Coulomb balance, and of compensating for this error, or at least a part of it, after learning it.

[0033] The idea is to consider that the state of charge learned from the open-circuit voltage is accurate; This measurement allows us to evaluate the error made by the Coulomb estimator. Thus, during the open-circuit voltage measurement, the battery management system compares the state of charge estimated by the Coulomb method with the new state of charge obtained from the open-circuit voltage. The difference is measured. From this, the error made during the Coulomb integration is deduced and compensated for (at least partially) in subsequent Coulomb integration phases.

[0034] Thus, a battery will have its state of charge better and better estimated as it is used.

[0035] The invention will also allow the estimator to be adapted to the aging of the battery and its components (if it were to have an effect on the estimation of the state of charge).

[0036] The invention applies to the state-of-the-art device in LFP technology. The battery management system uses voltage information to estimate the state of charge, whenever conditions allow (in particular the area where the voltage varies sufficiently with the state of charge), and after a sufficient relaxation time.

[0037] Apart from these situations, the battery management system estimates the state of charge by integration and Coulomb balance: SOC(t) = SOC(t-At)-^o^ SOC(t) is the current state of charge; SOC (t- A t) is the preceding approximate state of charge; A t is the difference between a current time and a previous time; I is the current in amperes; This is the capacity in As

[0038] The invention then consists of: each time the battery management system has the opportunity to accurately estimate the state of charge from the open-circuit voltage, the battery management system also calculates the difference between the state of charge estimated just before (by the Coulomb method) and the new, more accurate state of charge value. This difference is called ASOC.

[0039] The Coulomb calculation incorporates a correction factor a, such that: SOC(t) = SOC(tA f) - °Ù SOC(t) is the current state of charge; SOC(t- A t) is the preceding approximate state of charge; A t is the difference between a current time and a previous time; a is the correction factor; I is the current in amperes; This is the capacity in As

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051] The explanation is that if we consider that the current measurement is affected by a systematic error during the measurement period by Coulomb integration. Let us also consider that this error consists of a shift error d (or "offset error" in English) and a gain error g. The calculation of the state of charge in the Coulomb integration phase can thus be written as: SOC(t) = SOC( t- A / ) - IM the actual current. We obtain: ( r - A 0 = SOC ( r ) - ( 0 If we take the example of a cycle of several discharges and several charges that restore the battery to the same state of charge (for example, 100%), and without any shift error; then we can easily deduce from the previous equation that: From this we can deduce what is quite intuitive, namely that the ASOC error is solely related to the offset d. The gain errors g compensate each other between the charging and discharging phases, whereas the offset error d does not compensate, and is constantly added as the integration progresses. The ASOC error accumulated over the Coulomb integration period is then simply directly proportional to the lag d, to the integration time (and inversely proportional to the capacity of the cells). For example, for an integration lasting 100 h, a 50 mA offset error, of 100 Ah cells, generates a state of charge error of 50 mA * 100h / 100 Ah, or 5%. Since the battery management system knows the integration time, the battery management system is therefore able to estimate the offset error each time it is able to perform an accurate measurement by the open-circuit voltage. The invention consists of reintroducing the shift error into the Coulomb calculation formula. Thus, the calculation becomes, according to the aforementioned formula, incorporating the correction factor. In practice, in order to ensure the stability of the convergence process, it is preferable to reinject only part of the gap, for example half of the observed gap. Thus, the parameter can be either the offset error, or preferably half of the offset error. Example: For a city car belonging to the applicant, starting at 80% charge level, the vehicle is used and reaches 30% charge level. After that, it is recharged to 80% charge level. The battery management system only uses Coulomb integration. The current measurement has two errors: a gain of 0.5%, and a shift of 100 mA.

[0052] At the end of a motorway journey, the error mainly stems from the gain, but after recharging, the gain itself has balanced out, leaving only the offset error. This error accumulates over the long term.

[0053] When the battery management system is able to use the open-circuit voltage (for example, at a full load greater than 100%), it can measure the previous underestimation, here at 12% in this example. The battery management system also knows the time since the last open-circuit voltage measurement at 100%: here, 175 hours in this example. Therefore, the estimated offset error is ^SOC = d^

[0054] Thus, d = ^SOC^ Ad

[0055] In this case, the offset is 100 mA

[0056] In the following cycles, a measurement with half the offset will be used.

[0057] The invention further relates to the corresponding estimation method and program.

Claims

Claims

1. A device for estimating the state of charge of a battery connected to the device, the battery comprising battery cells having a positive electrode comprising lithium iron phosphate, the estimating device comprising: - correction means (M1) determining a correction factor based on the difference between an accurate state of charge estimated based on the no-load voltage, and the corresponding approximate state of charge estimated based on a previous approximate state of charge, the cell current and the cell capacity; - estimation means (M2) estimating the current state of charge based on a previous approximate state of charge, the cell current and the cell capacity, including said correction factor.

2. Estimating device according to claim 1, characterized in that the estimating means estimates the current state of charge on the basis of the formula: SOC(t) = SOC(t- A o - SOC( / ) is the current state of charge; SOC (t- A / ) is the previous approximate state of charge; A / is the difference between a current time and a previous time; a is the correction factor; I is the current in amperes; C is the capacity in As

3. Estimating device according to any one of claims 1 to 2, characterized in that the correction factor is a current offset error between the precise state of charge estimated on the basis of the no-load voltage, and the corresponding approximate state of charge.

4. Estimating device according to any one of claims 1 to 2, characterized in that the correction factor is half of a current offset error between the precise state of charge estimated on the basis of the no-load voltage, and the corresponding approximate state of charge.

5. Estimation device according to claim 4, characterized in that the correction factor is determined by means of the formula a - 4 -1 * A SOC '^ where a is the correction factor; d is the offset error; A t is the difference between a current time and a previous time; A SOC is the difference in state of charge between the current time and a previous time.

6. A battery management system comprising an estimation device according to any one of claims 1 to 5.

7. A motor vehicle comprising: - a traction battery comprising battery cells having a positive electrode comprising lithium iron phosphate; and - an estimation device according to any one of claims 1 to 5, connected to said traction battery.

8. A method for estimating the state of charge of a battery comprising battery cells having a positive electrode comprising lithium iron phosphate, by means of an estimation device according to any one of claims 1 to 5, the estimation method comprising the following steps: - a correction step (El) in which a correction factor is determined on the basis of the difference between the precise state of charge estimated on the basis of the no-load voltage, and the corresponding approximate state of charge estimated as a function of a preceding approximate state of charge, the cell current and the cell capacity; - an estimation step (E2) in which the current state of charge is estimated as a function of a preceding approximate state of charge, the cell current and the cell capacity, including said correction factor.

9. Estimation method according to claim 8, characterized in that in the estimation step, the current state of charge is estimated on the basis of the formula: SOC(t)=SOC(t-At)-^^ SOC{ t) is the current state of charge; SOC( t- A t) is the previous approximate state of charge; A t is the difference between a current time and a previous time; a is the correction factor; I is the current in amperes; C is the capacity in As

10. A computer program comprising program code instructions for carrying out the steps of the estimation method according to any one of claims 8 to 9, when said program works on a computer.

Images