KR20140071060A - Methods and apparatus for online determination of battery state of charge and state of health - Google Patents

Abstract

Provided are a method and an apparatus for online determination of a battery charging state (SoC) and a battery health state (SoH) on a platform providing a dynamic charge and discharge environment. A pause open circuit voltage (OCV) can be predicted online along with a measured terminal voltage, current, and temperature by using a dynamic model. Then, the SoC and the SoH can be determined from the predicted OCV. The apparatus and the method can a) block a battery system from a service, b) wait during predefined break period, or c) predict the SoC and the SoH of the battery without depolarizing the battery.

Description

[0001] METHOD AND APPARATUS FOR ONLINE DETERMINATION OF BATTERY STATE OF CHARGE AND STATE OF HEALTH [0002]

The present invention relates to a method and apparatus for determining a state of charge (SoC) and a state of health (SoH), and more particularly, to a method and apparatus for determining a SoC and SoH of a battery in a dynamic charging and discharging environment. And an apparatus and method for online judgment.

Rechargeable batteries are an important component of recently emerging microgrid electric vehicles (EVs), plug-in hybrid electric vehicles (PHEVs) and more electric aircraft (MEA) systems. Such a platform requires frequent charging and discharging of the battery. For reliable operation and to conserve battery life, it is essential that accurate knowledge of the battery's state of charge (SoC) and effective battery health (SoH) is known. Effective battery capacity is a leading indicator of SoH.

The prior art provides several SoC and SoH decision techniques. However, this technique can be applied only when the battery is in an off-line mode, i.e., in a laboratory environment, or when it is not used to support charge and discharge functions within the host environment. Due to the unusual behavior of the battery under dynamic charging / discharging conditions occurring in on-line mode, this technique is not applicable or can not provide accurate results.

The most basic technique for determining an SoC is based on open circuit voltage (OCV) or dormant voltage measurements. The OCV is generally defined as the battery terminal voltage after being idle for a predetermined time of at least 30 minutes to several hours without load or charge. For many lithium-ion and other battery chemistries, OCV varies with SoC and can not be used to calculate SoC. Figure 1 provides an example of the OCV-SoC interrelationship for the GS Yuasa 28V, 40Ah lithium ion battery at 0, 25 and 45 ° C. This figure shows a nearly linear relationship between SoC and OCV in the approximately 10% SoC to 100% SoC range. OCV versus SoC graphs are generally available from battery / cell suppliers, or can be constructed through laboratory testing. Using this approach, better algorithms have been developed than +/- 5% SoC accuracy. However, this approach is unsuitable for on-line SoC determination in a dynamic charge / discharge environment where a dormant battery voltage is rarely available due to polarization phenomenon. However, the polarization phenomenon is illustrated in Figures 2A and 2B showing the battery response to a repetitive 10A load pulse. The magnified view depicted with respect to the battery voltage shows a sudden jump in voltage and a much slower voltage recovery following the removal of the load, which requires a long wait time to capture the idle battery voltage. Waiting time is usually not practical in an online environment.

Battery capacity decreases gradually over time and usage. If the capacity of the battery is reduced by 50% in 100% SoC, it will have only half the energy when it is used for the first time. Therefore, the SoC measurement alone is not sufficient to ensure efficient operation of the battery in the system. Battery capacity, which is the main indicator of SoH, should also be monitored.

In the prior art, there are methods for predicting battery capacity, but all of these methods can only be implemented off-line and require a waiting time for the battery to reach equilibrium. As described in U.S. Patent No. 7,576,545, the total capacity of the battery can be determined through partial charge / discharge. The method starts with a known SoC state, and after a known amount of energy is added or subtracted, the idle open-circuit voltage of the battery is measured and a new SoC is calculated. Then, the total capacity C _full can be obtained by correlating the charge / discharge energy [ Delta ] E with the change in the SoC ([ Delta] SoC ) by Equation (1).

[Equation 1]

C _full * DELTA SoC = ΔE

Here ,? SoC = | SoC _after - SoC _before | to be. However, after partial charge / discharge, the idle open circuit voltage needs to be obtained via depolarization / predefined downtime and therefore interrupts system operation. This is not practical when the battery is online.

As can be seen, there is a need for on-line technology to determine SoC and SoH of a battery.

In one aspect of the present invention, a system for determining a state of charge of a battery and a state of health includes a measurement unit having a voltage sensor, a current sensor, and a temperature sensor and taking a measurement value from the battery; A power source connected to the battery via a power switch; A load connected to the battery via a load switch; And a computer having a control card for receiving a measurement value from the measurement unit and opening and closing the power switch and the load switch, and calculating an open circuit voltage on-line without requiring a battery downtime.

In another aspect of the present invention, a method for online determination of a battery open circuit voltage without requiring a battery downtime includes: charging or discharging the battery; Calculating the open circuit voltage using V _OC = V _BAT + ? V _R + ? V _P ; And determining a state of charge using the calculated open circuit voltage and the battery temperature, wherein V _BAT is the battery terminal voltage,? V _R is the voltage drop due to the battery resistance,? V _P is the polarization state Resulting in a voltage drop.

In another aspect of the present invention, a method of determining a health condition of a battery includes: estimating a total capacity of the battery by charging or discharging the battery; Calculating an open circuit voltage using V _OC = V _BAT + ? V _R + ? V _P ; Determining a state of charge using the calculated open circuit voltage and battery temperature; Correlating the charge or discharge energy with the variation in the charge state to estimate the total battery capacity; Tracking the total capacity of the battery as a function of time; And determining whether the total capacity of the battery has decreased by a predetermined amount, where V _BAT is the battery terminal voltage,? V _R is the voltage drop due to the battery resistance, and? V _P is the voltage drop due to polarization Voltage drop.

These and other features of the present invention will be better understood by reference to the following drawings, specific details for carrying out the invention and claims.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a graph showing a paused battery voltage vs. battery charge state, as is known in the art;
2A is a graph showing the battery terminal voltage under dynamic load conditions;
Figure 2b is a detail portion of the graph of Figure 2a;
3 is a graph showing voltage recovery versus time due to polarization;
4 is a chart showing an exemplary look-up table;
5 is a schematic diagram illustrating a system for determining SoC and SoH in accordance with an exemplary embodiment of the present invention;
Figure 6 is a schematic diagram illustrating a battery model with a resistor ( Rs ) in series with two RC branches in accordance with an exemplary embodiment of the present invention;
7 is a flowchart illustrating a method of determining a battery SoC in accordance with an exemplary embodiment of the present invention;
8 is a flowchart illustrating a method of measuring battery capacity according to an exemplary embodiment of the present invention; And,
9 is a flowchart illustrating a method of determining a battery SoH according to an exemplary embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following detailed description is a best mode contemplated for carrying out the exemplary embodiments of the invention. The description is not to be taken in a limiting sense, and is only for the purpose of illustrating the general principles of the invention, since the invention is best defined by the appended claims.

Various progressive features, which may be used independently of each other or in combination with other features, are described below.

Broadly, embodiments of the present invention provide an online method and apparatus for determining the state of charge (SoC) and state of health (SoH) of a battery on a platform that provides a dynamic charging and discharging environment. The dormant open circuit voltage (OCV) can be predicted online using the battery dynamic model along with the measured terminal voltage, current, and temperature. Next, SoC and SoH can be determined from this predicted OCV. The apparatus and method comprise: a) blocking the battery system from service; b) waiting for a predefined pause time; Or c) predict the SoC and SoH of the battery in real-time without the need to depolarize the battery.

The method of the present invention can be implemented using the components shown in FIG. 5, as described above. The battery monitoring system 10 may include a single cell or a battery 12 that may include multiple cells in series or parallel. The system 10 may further include a fixed or programmable power source that is used to partially or fully charge the battery 12, which is connected across the battery 12 to a controllable switch 16. The system 10 also includes a resistive load or programmable electronic load 18 for discharging the battery 12 into a constant current or predefined current configuration coupled to the battery 12 by a controllable switch 20 . The measurement unit 22 may include an ammeter or current sensor 24, a voltmeter or voltage sensor 26 and a temperature sensor 28 to sense the measured value and provide a data acquisition system (DAQ) 30 ). The system 10 includes a data acquisition system 30, a non-volatile memory (NVM) 34 for storing the lookup table and data, a processor 36 for executing the method, and a display 38). ≪ / RTI > The data acquisition system 30 may include a sample and hold circuit, an A / D converter, and a non-spreading filter (not shown) to remove sensor noise. The computer 32 can control the charging and discharging cycles of the battery 12 by controlling the switches 16 and 20 via the control card 40. [

An exemplary embodiment of the present invention provides a method for calculating a dormant open circuit voltage on-line. The on-line method can use the battery dynamic model together with the measured terminal voltage, current and temperature to predict the open-circuit voltage. The battery model may be any linear / nonlinear equivalent electrical model representing the battery characteristics. The battery model may include resistance, capacitance, and inductance in series with a slave voltage source. Model parameters can be identified through pulse current charge / discharge. To develop the model, testing can be performed at different temperatures for various charge / discharge currents. The battery open circuit voltage can be calculated on-line using Equation 2 associated with the model. The SoC can be predicted through a lookup table / algebraic equation using the calculated OCV and temperature.

&Quot; (2) "

V _OC = V _BAT + ? V _R + ? V _P

Here, V _BAT is the battery terminal voltage ,? V _R is the voltage drop due to the battery resistance, and ? V _P is the voltage drop caused by the polarization phenomenon.

Referring to FIG. 6, these values may be calculated using an electrical model representing, for example, the characteristics of a lithium ion battery. Battery model may include two RC branches in series with the resistance (R _s) as shown in Fig. Each RC branch can contain a parallel combination of resistance and capacitance.

The battery terminal voltage (V _BAT ) can be measured using a voltage sensing device that can be connected across the battery terminals.

Voltage drop caused by a battery internal resistance (Δ V _R) can be calculated using equation (3).

&Quot; (3) "

Δ V _R = I * _{R s}

Here, R _s is a battery internal resistance, I is the current flowing through the battery terminals.

The polarization phenomenon can be represented using two RC branches in series as shown in FIG. 6, and the drop due to this phenomenon can be calculated using Equation (4).

&Quot; (4) "

Here, R _st and C _st are elements of a first RC branch showing a short-time polarization phenomenon after the battery is kept in a rest state, and R _lt and C _lt are long-term polarization phenomena after the battery is kept in a rest state Lt; RTI ID = 0.0 > RC < / RTI >

Equation (2) can be written as

&Quot; (5) "

The initial battery parameter can be predicted from the battery response immediately after the load is removed. As discussed in the previous section, a sudden jump in voltage is due to the battery internal resistance ( R _S ) and can be predicted from equation (6).

&Quot; (6) "

The slow voltage recovery due to the polarization phenomenon can best be fitted to Equation (7) as shown in Fig.

&Quot; (7) "

Battery polarization parameters _{(R st, C st, R} lt, C lt) can be calculated as follows from the coefficients a, b, c and d in the equation (7).

&Quot; (8) "

&Quot; (9) "

&Quot; (10) "

&Quot; (11) "

This process can be repeated with different SOC levels for different loads at different operating temperatures. Using any circuit simulator, the best fit parameters can be identified and authenticated through circuit simulation. These values for the new 28V, 40Ah lithium-ion battery are as follows:

R _s = 0.01388?;

R _st = 0.01394?;

C _st = 11031.218 F;

R _lt = 0.01531?; And

C _lt = 4662154.6 F.

These battery parameters vary with battery usage and aging. These parameters can be predicted at regular time intervals by monitoring and storing battery data in real time for different system loads.

Embodiments of the present invention also disclose an online model update method that results from fluctuations in battery parameters from cycling and aging. For this purpose, battery characteristics can be performed on-line by monitoring the response to a suitable system load; Voltage, current and temperature are parameters used in this process.

Referring to FIG. 7, a method 50 for determining the state of charge of a battery is described in accordance with an exemplary embodiment of the present invention. In step 51, the OCV is measured and the initial SoC can be determined. At this point, the polarization may not be a problem because the battery is already at rest. In step 52, charging / discharging of the battery starts, and in step 53, the battery terminal voltage, charge / discharge current and temperature can be measured. In optional step 54, the model parameters may be updated with respect to the current state, temperature or lifetime of the battery. In step 55, the OCV can be calculated using Equation 2 above. In step 56, the SoC may be determined using the calculated OCV and temperature from step 55 through a look-up table or through an algebraic equation. At step 57, this SoC can be stored in memory, displayed to the user, or returned to do.

Referring to Figure 8, a method 60 for determining battery capacity in accordance with an exemplary embodiment of the present invention is described. In step 61, the OCV is measured and the initial SoC is determined. At this point, the polarization may not be a problem because the battery is already at rest. In step 62, the charging / discharging of the battery is started, and in step 63, the battery terminal voltage, charging / discharging current and temperature can be measured. If the method 60 is charging the battery, the method 60 may follow the left side of the flowchart. If the method 60 is discharging the battery, the method 60 may follow the right side of the flowchart. In optional step 64, the model parameters may be updated with respect to the current state, temperature or lifetime of the battery. In step 65, the OCV can be predicted using Equation 2 above. In step 66, the SoC may be determined through the predicted OCV. In step 67, the charge / discharge energy used is correlated with the variation in the SoC, and the total battery capacity is estimated. Additional details of step 67 can be found in U.S. Patent No. 7,576,545, the contents of which are incorporated herein by reference.

Referring to FIG. 9, a method 70 for determining battery SoH is described. The method 70 may use the predicted total battery capacity, as determined in the method described above with reference to FIG. The method 70 includes the step of tracking 71 the total capacity of the battery as a function of time and the step 72 of performing a trend analysis of the total capacity of the battery over time to predict the end of life of the battery . If the total capacity of the battery decreases by a predetermined amount, then the method 70 determines that the battery has reached its end of life if the capacity of the battery has been reduced by at least the predetermined amount, otherwise the battery is still good And a judgment point 73 at which the judgment can be continued.

The total available capacity of the battery is a key parameter that determines the life of the battery. The method described above may include a mechanism to track the total capacity of the battery as a function of time and to perform a trend analysis on the total capacity of the battery to predict the end of life of the battery and suggest battery replacement.

The methods of the present invention solve the problem of predicting the SoC and SoH of a battery in a real-time manner without a) blocking the battery system from service, b) waiting for a predefined pause time, or c) do.

It is, of course, to be understood that the foregoing is directed to the exemplary embodiment of the invention and that modifications may be made without departing from the spirit and scope of the invention as set forth in the claims.

Claims (10)

A measurement unit 22 having a voltage sensor 26, a current sensor 24 and a temperature sensor 28 and taking a measurement value from the battery 12;
A power source 14 connected to the battery 12 through a power switch 16;
A load 18 connected to the battery 12 via a load switch 20; And
And a control card (40) for opening and closing the power switch (16) and the load switch (20) by receiving a measured value from the measuring unit (22) Lt; RTI ID = 0.0 > 32 <
/ RTI >
A system for determining battery charge status and health status.
The method according to claim 1,
The open circuit voltage
V _OC  =V _BAT  +Δ V _R  +Δ V _P
Lt; / RTI >
Here, V_BATIs the battery terminal voltage,
ΔV_RIs a voltage drop due to battery resistance,
ΔV_PWhich is a voltage drop due to the polarization phenomenon,
A system for determining battery charge status and health status.
3. The method of claim 2,
? V _P

≪ / RTI >
Where R _st and C _st are elements of the first RC branch that exhibit short-term polarization after holding the battery in a rest state,
RTI _{ID = 0.0} & _gt; R & _lt; / _RTI & _gt; and C & _lt; RTI _{ID = 0.0} & gt; _lt & _lt; / _RTI & _gt; are elements of a second RC branch exhibiting long-
A system for determining battery charge status and health status.
The method according to claim 2 or 3,
Further comprising a data acquisition system (30) for collecting data from the measurement unit,
A system for determining battery charge status and health status.
5. The method according to any one of claims 2 to 4,
Further comprising a non-volatile memory (34) for storing a look-up table and data.
A system for determining battery charge status and health status.
A method for online determination of a battery open circuit voltage without requiring a battery downtime,
Charging or discharging the battery (12);
The open-circuit voltage
V_OC = V_BAT + ΔV_R + ΔV_P
; And
Determining the state of charge using the calculated open circuit voltage and battery temperature
Lt; / RTI >
Here, V_BATIs the battery terminal voltage,
ΔV_RIs a voltage drop due to battery resistance,
ΔV_PWhich is a voltage drop due to the polarization phenomenon,
How to make online determination of battery open circuit voltage.
The method according to claim 6,
Up table is used to determine the state of charge of the battery using the calculated open circuit voltage and the battery temperature,
How to make online determination of battery open circuit voltage.
8. The method according to claim 6 or 7,
Further comprising the step of initially measuring an open-circuit voltage of the battery in a dormant state to determine an initial battery charge state,
How to make online determination of battery open circuit voltage.
9. The method according to any one of claims 6 to 8,
Further comprising periodically updating model parameters for current, charge state, temperature and battery life,
How to make online determination of battery open circuit voltage.
10. The method according to any one of claims 6 to 9,
Correlating the charge or discharge energy with the variation in the charge state and estimating to obtain the total battery capacity,
How to make online determination of battery open circuit voltage.

Images