TECHNICAL FIELD

This invention relates to a method for calibrating a current sensor arranged for measurement of current from or to a battery system provided with one or more battery cells.

BACKGROUND OF THE INVENTION

Electric energy storage systems in the form of battery packs/systems are becoming more and more common and important worldwide in various stationary and mobile applications. A battery pack typically includes a plurality of battery cells, such as of Li-ion type, connected in series and/or in parallel. The battery pack further typically includes a battery management system (BMS) configured to control the battery pack, such as to control charging and discharging of the battery pack and determine state of charge (SOC) and state of health (SOH).

The battery pack typically includes a current sensor connected to the BMS for measuring inflow of charge (charge current) and outflow of charge (discharge current) to and from the battery pack. The current sensor may also be used to detect too high currents, in which case the BMS may take some action such as turning off the entire system, and to determine total SOC (together with measured cell voltage and temperature).

To allow for the BMS to control the battery pack properly, the input from e.g. the current sensor must be reasonably correct. Well-calibrated current sensors may be obtained by high accuracy manufacturing or pre-calibration of individual sensors, but such sensors are relatively expensive. An alternative is to use more simple and less costly sensors and manually calibrate a certain current sensor to be arranged in a certain battery system, but this is rather time-consuming and re-calibration might be needed so the time-consuming procedure might need to be repeated. There is thus a need for providing a calibrated battery pack current sensor in a way that is efficient with regard to both cost and time. JP2015224975A discloses a discharge current detection device intended to correct a current value detected by a current sensor and accurately detecting a current value of a charge / discharge current even when an electric load has an unknown current consumption value. Besides that what JP2015224975A discloses is somewhat difficult to understand in full, the method of JP2015224975A requires that the battery is connected to a load, such as a vehicle, computer or mobile terminal, and it also involves calculations of a second charge change amount based on the charge rate change amount of the battery. Since it is useful to calibrate the current sensor also when the battery pack is not connected to a load, for instance in association with assembling of the battery pack, and since involving not only the charge change but also the rate of the change makes the procedure complicated, the method of JP2015224975A is not suitable for solving the problems addressed in this disclosure. SUMMARY OF THE INVENTION

The invention concerns a method for calibrating a current sensor arranged for measurement of current from or to a battery pack/system provided with one or more battery cells, the method comprising the steps of: providing information on a relationship between a voltage and a charge level of the battery system; charging or discharging the battery system so as to change the charge level and thereby the voltage of the battery system; measuring during said charging or discharging at least a first voltage and, at a later point in time, a second voltage; obtaining, based on the relationship between the voltage and the charge level of the battery system, a total reference charge corresponding to the change of voltage from the first voltage to the second voltage; measuring the current, using the current sensor, during said charging or discharging; obtaining a total measured charge based on the current sensor measurement; and calibrating the current sensor based on a difference between the total reference charge and the total measured charge.

The step of “providing information on a relationship between a voltage and a charge level of the battery system” means that information is provided on the voltage (V) as a function of charge level (C, Ah, %, etc.). With discharge level on the x-axis (i.e. full charge at x=0) and voltage on the y-axis this function typically has a general negative slope with a relatively linear mid-section and increased (negative) slope close to the end points, i.e. close to fully charged and fully discharged battery system (or battery cell). The principle of this relationship is well known to a person skilled in the art. In the method of this disclosure, the information on this relationship is typically stored in a memory accessible for the BMS so that the BMS can use this information in calculations.

At least for high-quality battery cells, an accurate and reliable charge level- voltage relationship can be obtained from the manufacturer of the cells. Alternatively, these data can be obtained from measurements of the particular type of cell to be used. Information for individual cells can then be adapted to the particular battery system in question. As an example, the battery system may contain 36 Li-ion cells connected in parallel and forming a battery system having a capacity of 105 Ah and a voltage of around 3.6 V (depending on charge level).

The charge level-voltage relationship may change over time due to aging of the battery cells, but for new and relatively new battery cells the data from the manufacturer (or from measurements) can be used. The method of this disclosure is therefore particularly well adapted for being used in connection with assembly of new battery packs.

The step of “charging or discharging the battery pack so as to change the charge level and thereby the voltage of the battery pack” means that the battery pack is charged or discharged to some degree so that the voltage changes according to the known relationship.

In the step of “measuring during said charging or discharging at least a first voltage and, at a later point in time, a second voltage” the purpose is to obtain at least two values for the voltage on the known charge level-voltage curve so that the corresponding change in charge level can be obtained. The number of voltage measurements may of course be more than two to reduce measurement uncertainties. Handling of several voltage data points may include parameter fitting or regression analysis. Further, charging/discharging may be maintained over the entire level-voltage curve (from fully charged to fully discharged or vice versa) or over only a portion of the curve. In principle it is sufficient to measure only a first voltage and a second voltage, where the second voltage is measured at a later point in time during (dis)charging so that the charge level has changed.

It can then be obtained, based on the relationship between the voltage and the charge level of the battery pack, a total reference charge corresponding to the change of voltage from the first voltage to the second voltage. This total reference charge is thus the assumed true total charge that has flown into or out from the battery pack during charging/discharging based on data from e.g. the battery cell manufacturer.

By measuring the current, using the current sensor, during said charging or discharging, it can be obtained a total measured charge based on the current sensor measurement. The current sensed by the current sensor is integrated over the measurement period (e.g. the time period between the measurement of the first and second voltages or the time period between the first and last voltage measurements in a series of voltage measurements) so as to obtain the total measured charge. By comparing the total reference charge with the total measured charge, information is provided on whether the current sensor measures the current accurately and, if not, to what degree the current sensor erroneously measures the current. If needed, the current sensor is then calibrated. A threshold value may be set so that calibration is performed only when the fault in accuracy exceeds (an absolute value of) that threshold value.

The step of “calibrating the current sensor based on a difference between the total reference charge and the total measured charge” may involve adjusting a gain of the current sensor, possibly based on a calculated correction factor.

The BMS, i.e. a control circuitry configured to control the battery system, is preferably configured to carry out the steps of the above method. The BMS is thus arranged to e.g. control charging/discharging as well as voltage and current measurement, handle signals received from various sensors, make calculations, provide (or obtain) the charge level-voltage relationship from a memory, calibrate the current sensor, etc.

The invention also relates to a battery system comprising one or more battery cells, a current sensor and a control circuitry in the form of a battery management system, BMS, configured to control the battery system, wherein the control circuitry is configured to carry out the method according to above.

The invention also relates to a control circuitry in the form of a battery management system, BMS, configured to control a battery system comprising one or more battery cells and a current sensor, wherein the control circuitry is configured to carry out the method according to above.

BRIEF DESCRIPTION OF DRAWINGS

In the description of the invention given below reference is made to the following figure, in which:

Fig. 1a is a schematic block diagram illustrating a battery system.

Fig. 1b is a schematic block diagram illustrating a battery system. Fig. 2 is a flowchart depicting an example method.

Fig. 3 is a graph illustrating battery cell characteristics.

DESCRIPTION OF EXAMPLE EMBODIMENTS OF THE INVENTION Fig. 1a is a schematic block diagram illustrating a battery system 100 comprising in this case one battery cell 101, which may be a lithium-ion battery cell. The battery cell 101 has known cell capacity characteristics, e.g. charging and discharging characteristics, Knowing the cell capacity characteristics for a battery cell 101, e.g. voltage versus charge data, it should be understood that this data may be combined to provide corresponding data for a plurality of connected battery cells 101 of a battery module 108 (not shown in fig. 1a, but in fig. 1b), i.e. corresponding voltage and charge data may be achieved on a battery module level. A battery’s cell’s capacity may be described as the amount of electric charge which the battery cell 101 can deliver at a rated voltage. Cell capacity can be expressed in e.g. ampere hours (Ah).

The battery system 100 comprises a current sensor 103 connected to the battery cell 101. The current sensor 103 is capable of measuring inflow and outflow of charge per time unit (i.e. current) to and from the battery cell 101. Integrating current over time yields charge.

The current sensor 103 may be for example a shunt sensor, also referred to as an ammeter shunt, a current transformer, Rogowski coil, magnetic-field based transducer etc.

The battery system 100 further comprises a control circuitry in the form of a battery management system 105 connected to the current sensor 103 via a first link and to the battery cell 101 via a second link 106 that may be a communication bus such as e.g. a RS485 type (multidrop) communication bus, a Controller Area Network (CAN) bus, etc.. The battery management system 105 is adapted to manage the whole battery system 100, including calibration of the current sensor 103. The battery management system 105 coordinates when and how the battery system 100 may be used by an application, such as a vehicle or a machine. In other words, the battery management system 105 functions as a control unit controlling the battery system 100 and its interaction with the application it is supposed to provide electric power to. Thereby, the battery management system 105 communicates with and receives information from the current sensor 103 and the at least one battery cell 101.

The battery management system 105 uses measurements from the current sensor 103 for various purposes, such as for controlling charging and discharging of the battery system, for determining State Of Charge (SOC) and for calibrating the current sensor.

The battery management system 105 may be an electronic control board, and the control signals to and from the battery management system 105 may be wireless.

Fig. 1b illustrates the battery system 100 comprising a plurality of battery cells 101, which may be serially connected or connected in parallel. The plurality of battery cells 101 may be comprised in one or more battery modules 108. The battery module 108 comprises one or a plurality of battery cells 101 which form a storage unit for electrical energy of the battery module 108. Each battery module 108 may comprise for example 36, 52 or 72 battery cells 101, corresponding to 105, 150 or 210 Ah connected in parallel. Each battery module 108 may comprise any other n number of battery cells 101, where n is a positive integer.

The battery module 108 may be referred to as a battery pack, a pack comprising a plurality of battery cells 101 etc. Also the entire battery system 100 may be referred to as a battery pack. The battery module 108 may be described as an energy storage device, which is comprised of electrically connected battery cells 101 or modules. It may incorporate a protective housing and be provided with terminals or other interconnection arrangement. It may include protective devices and control and monitoring.

Each battery module 108 may comprise a battery management unit (not shown) respectively, adapted for determining voltage and temperature of the at least one battery cell 101 within the battery module 108. The battery management unit which may be comprised in the battery module 108 may be an electronic subsystem which may measure battery module voltages and temperatures. The battery management unit may communicate with the battery management system 105 or with a part of the battery management system 105. The communication may be done via a communication bus, wired or wireless.

The battery system 100 may comprise other units such as e.g. a fuse, a switch, a power switch, a relay, a memory, input interfaces, output interfaces, busses, temperature sensors, cooling system heating system, clock, power connector etc., which are not illustrated in fig. 1a and fig. 1b.

The battery system 100 may comprise an interface towards a charger. The charger may be an integrated charger which is integrated in the battery system 100 or it may be an external charger adapted to be connected to the battery system 100.

An example of the method will now be described with reference to fig. 2 that shows a flow chart illustrating the method performed by the battery management system 105. The method comprises at least one of the following steps to be performed by the battery management system 105, which steps are performed in any suitable order than described below: Step 301

The at least one battery cell 101 is charged to a first charge level, e.g. 100% charge, 90% charge, full charge etc. This may be described as In other words, the voltage of the at least one battery cell 101 at OCV is equal to the voltage when the at least one battery cell 101 is charged to the first charge level.

Step 302 Zero current integration may be performed. This may be described as:

C _int = 0

Step 303

In this step, the charge integration may be performed. This may be described as:

, where ^i(t ₎ is the current at time t.

Step 304

In this step, the battery management system 105 may check if the at least one battery cell 101 is below the first charge level, e.g. if it is fully charged or empty, i.e. it checks if . Here it may be assumed that i ^(t) has been low for a predetermined amount of time. This may ensure that is close to , i.e. to the voltage of the at least one battery cell

101 at OCV.

Step 304 In this step, the correction factor ^K is calculated. The correction factor ^K may be calculated as follows: is the integrated charge. is the cell capacity of the at least one battery cell 101 at

Beginning of Life (BOL).

Calculation of the correction factor ^K may be applied to a series of N charge cycles or discharge cycles in order to minimize any type measurement error. N may be any positive integer, e.g. 10. This may be be done with a 1 pole recursive filter like

, where alfa is between 0 and 1 depending on how fast the new values of ^K should be weighted.

When N charge or discharge cycles have been retrieved, the value, i.e. the accumulated correction factor or the total correction factor, may be used to adjust the current sensor gain ^G . Assuming that the measured current

When inserting the correction factor , the calibrated current at a time instance t may be as follows: G is the sensor gain is the total or accumulated correction factor. is the current measured by the current sensor 103. t is the time. is the callibrated current.

The method described above will now be described with an example where there are 10 number of points in the discharge cycle, i.e. N=10 as an example. 10 measurements are taken in a discharge cycle. The example is equally applicable to 10 number of charge cycles. The method comprises at least one of the following steps 1 )-9), which steps may be performed in any suitable order than described below:

1 ) Charge the at least one battery cell 101 to a first charge level.

Assume

2) The following fucntion is defined:

3) Set . The integration is set to zero when the at least one battery cell 101 is charged to the first charge level.

4) . In this step, the at least one battery cell 101 is used, i.e. discharged during a normal operation.

5) If , then set

6) If ^k≤N , then step 4) is performed.

7) Calculate where ^f(x) is the function from voltage to discharged capacity, i.e. to the second charge level, as seen in point D and C in fig. 4 (described below).

8) Calculate the Leastsquare fit:

9) Set the new The above described method may be combined with a requirement to use more than one cycle and thus slowly adjust the to the new data and become even more independent of e.g. measurement noise associated with the current sensor 103.

There may be one or more prerequisites for the present disclosure. These will now be described with reference to the graph illustrated in fig. 3 that illustrates an example of a charge capacity for at least one battery cell 101. The x-axis of fig. 3 represents the discharged charge measured in Ah. The y- axis of fig. 3 represents the cell OCV voltage measured in V.

A prerequisite may be that the at least one battery cell 101 may have a total cell charge capacity (Ctot) that is known and possibly with a known tolerance or accuracy. The total cell charge capacity, Ctot, may be defined as the amount of current, e.g. integrated current, from a battery cell 101 charged to the first charge level (e.g. full cell), see point F in fig. 3, to a battery cell 101 charged to the second charge level (e.g. empty cell), see the crossing between points A and B in fig. 3. As mentioned above, the first charge level is higher than the second charge level. The first charge level may be full charge and the second charge level may be zero charge, i.e. empty cell. When the cell voltage reaches the B point, then it is equal to OCV, VcellSecondChargeLevel, a new battery cell 101 has at this point discharged a known amount of integrated current, see point A, aka cell capacity at BOL. This accumulated or integrated current is known and fixed to a certain tolerance from the battery cell manufacturer. This is valid in both directions charge or discharge of the at least one battery cell 101.

Another prerequisite may be that the at least one battery cell 101 has a known OCV versus discharge characteristic for every point between a battery cell 101 at the first charge level and a battery cell 101 at the second charge level, and at BOL. For each position during a discharge cycle the battery management system 105 may obtain pairs of Voltages Vk and discharged integrated current Ck. It may be obtained so that the pairs are collected spread out over range from the first charge level to the second charge level of the at least one battery cell 101. For example, only one point (Vk, Ck) may be collected in each 1/10 of the discharge range. This method is valid in both directions, i.e. in both charge and discharge of the at least one battery cell 101.

The present disclosure takes advantage of using at least one battery cell 101 in the battery system 100, with known charge accuracy and discharging characteristics, when calibrating the current sensor 103 arranged in the battery system 100. The known charge accuracy and discharging characteristics are data available from battery cell manufacturers.

Fig. 3 illustrates an exemplifying discharging characteristic of at least one battery cell 101. The at least one battery cell 101 has known OCV versus discharged charge for every point between the first charge level to the second charge level, e.g. from a fully charged cell to an empty cell.

The cell charge capacity of the at least one battery cell 101 may be defined as the amount of charge from a battery cell 101 with a first charge level, see point F, to a battery cell 101 with a second charge level, see crossing between A and B. When the cell voltage reaches the B point then it is equal to the OCV Voltage (VcellSecondChargeLevel) of a battery cell 101 with the second charge level and at this point, a known amount of integrated current has been discharged, see point A, which amount is the battery cell capacity. This accumulated or integrated current, charge, is known and fixed to a certain tolerance from the battery cell manufacturer and represents the characteristics of the battery cell 101 as new, i.e. in BOL.

As the at least one battery cell 101 ages, inter alia by means of being repetitively discharged and charged, the cell characteristics change and/or deteriorate. Thereby, the known cell capacity characteristics have acceptable accuracy based on that the battery cells 101, i.e. the battery module, is defined as being in BOL. Hence, the at least one battery cell 101 may not have been discharged and/or charged more than x times, where x is any suitable positive integer.

As mentioned earlier, the known cell capacity characteristics may comprise battery cell voltage versus charge data for every point between a battery cell 101 at the first charge level and a battery cell 101 at the second charge level. In this case, a discharge or charge cycle of the at least one battery cell 101, independent of the state of charge of the at least one battery cell 101 at beginning and/or end of the cycle, is used for the calibration of the current sensor 103. This has the advantage of not having to fully discharge or fully charge the at least one battery cell 101 during the calibration of the current sensor 103. By way of example, referring to Fig. 3 again, a discharge cycle resulting in a remaining battery cell (OCV) voltage indicated by letter D gives a corresponding discharged charge capacity indicated by letter C.

The battery system 100 may be adapted to provide electric power to any suitable device which consumes electrical power. An electric vehicle may be an example of a device and may refer to any electrically propelled vehicle. The vehicle may refer to a vehicle at least partly supporting autonomous driving. The battery system 101 may form a charging station for an electric vehicle. The battery system 101 may be adapted for powering propulsion of a vehicle. The vehicle, which may comprise the battery system 100, may refer to any electrically driven vehicle, such as a land vehicle or a marine vessel. The battery system 100 may be adapted to provide electric power to vehicles, busses, boats, industrial applications, forklifts, trucks, Automated Guided Vehicles (AVG), machines such as e.g. mining machines, cleaning machines, construction machines, medical devices, power plants, wind turbines, communication equipment etc. In a less complex example, there is a method for calibrating the current sensor 103 arranged for measurement of current from or to the battery system 100 provided with one or more battery cells 101, the method comprising the steps of:

51 - providing information on a relationship between a voltage (V) and a charge level (Ah) of the battery system 100. Figure 3 shows a curve forming an example of such a relationship where a certain cell capacity/cell charge on the x-axis (with full charge of the battery system 100 at x=0 and increasing discharge with increasing x) corresponds to a certain voltage on the y-axis, and vice versa.

52 - charging or discharging the battery system 100 so as to change the charge level and thereby the voltage of the battery system. During charging one follows the curve of figure 3 in a direction to the left, leading to an increased voltage, and during discharging one follows the curve in a direction to the right, leading to a decreased voltage.

53 - measuring during said charging or discharging at least a first voltage and, at a later point in time, a second voltage. For instance, the battery system 100 can be charged or discharged between start and end points located at the two dashed lines in figure 3 where the first and second voltages are the measured (and differing) voltages at these two points.

54 - obtaining, based on the relationship between the voltage and the charge level of the battery system 100, a total reference charge corresponding to the change of voltage from the first voltage to the second voltage. The measured first and second voltages determine the start and end positions on the curve in figure 3 (i.e. where the solid horizontal lines intersect the charge curve) and the dashed lines point at the corresponding reference charge levels. The total reference charge for the battery system 100 in question, i.e. the total charge between the two reference charge levels, can then be calculated from the curve (i.e. from the relationship between the voltage and the charge level of the battery system 100).

55 - measuring the current, using the current sensor 103, during said charging or discharging. That is, the current sensor 103 measures the current to or from the battery system 100 during charging or discharging between the start and end positions for the charge level.

56 - obtaining a total measured charge based on the current sensor measurement. Integrating the signal from the current sensor during charging/discharging between the start and end positions for the charge level yields the total measured charge. If the total measured charge differs from the calculated total reference charge (by more than a certain value) it can be concluded that the current sensor is not properly calibrated.

57 - calibrating the current sensor based on a difference between the total reference charge and the total measured charge. This may be done by adjusting a gain of the current sensor. If the deviation between the total measured charge and the calculated total reference charge is less than a certain minimum threshold, it may be decided that the calibration is sufficiently good and that adjustment of the sensor gain is not necessary.

In this example the method steps above are carried out by the BMS 105.

It should emphasised that the steps of the methods above may be performed in another order than the order in which they appear.

The invention is not limited by the embodiments described above but can be modified in various ways within the scope of the claims.