U.S. patent application number 10/145938 was filed with the patent office on 2003-11-20 for state of charge algorithm for lead-acid battery in a hybrid electric vehicle.
Invention is credited to Ying, Ramona Y..
Application Number | 20030214303 10/145938 |
Document ID | / |
Family ID | 29400463 |
Filed Date | 2003-11-20 |
United States Patent
Application |
20030214303 |
Kind Code |
A1 |
Ying, Ramona Y. |
November 20, 2003 |
STATE OF CHARGE ALGORITHM FOR LEAD-ACID BATTERY IN A HYBRID
ELECTRIC VEHICLE
Abstract
A method and apparatus for determining the state of charge of a
lead-acid battery. The present invention includes steps to
determine a current-based state of charge while accounting for such
factors as current, temperature, charge efficiency, parasitic
losses and self-discharge. The present invention is capable of
measuring and/or calculating the open circuit voltage during and
after operation to determine a voltage-based state of charge while
compensating the open circuit voltage for its transient behavior
and accounting for the voltage shift caused by charge events. The
correlation of the open circuit voltage to the state of charge is
dependent on the battery's mode of operation. The present invention
is adaptive to battery aging by updating the battery resistance,
correcting coulomb errors, and accounting for capacity degradation
while in operation.
Inventors: |
Ying, Ramona Y.; (Rochester
Hills, MI) |
Correspondence
Address: |
CHRISTOPHER DEVRIES
General Motors Corporation
Legal Staff, Mail Code 482-C23-B21
P.O. Box 300
Detroit
MI
48265-3000
US
|
Family ID: |
29400463 |
Appl. No.: |
10/145938 |
Filed: |
May 15, 2002 |
Current U.S.
Class: |
324/426 ;
903/916; 903/917 |
Current CPC
Class: |
G01R 31/379 20190101;
Y02T 10/623 20130101; B60K 6/44 20130101; B60W 10/26 20130101; Y10S
903/917 20130101; G01R 31/374 20190101; Y10S 903/916 20130101; G01R
31/392 20190101; B60W 20/00 20130101; G01R 31/389 20190101; Y02T
10/62 20130101; B60W 20/13 20160101 |
Class at
Publication: |
324/426 |
International
Class: |
H02J 007/00 |
Claims
1. A method of determining the state of charge of a battery
comprising: integrating the charge going in and out of the battery
to determine a current based state of charge measurement;
compensating the current based state of charge measurement using a
Peukert factor; and measuring the open circuit voltage to determine
a voltage based state of charge measurement ; and compensating the
current based state of charge measurement based on the voltage
based state of charge measurement.
2. The method of claim 1 wherein said battery is a lead acid
battery.
3. The method of claim 1 wherein the Peukert factor is determined
using the average discharge current of the battery.
4. The method of claim 1 further comprising the step of determining
parasitic losses for the battery and compensating the current based
state of charge measurement based on the parasitic losses.
5. The method of claim 1 further comprising the step of determining
the self discharge of the battery and compensating the current
based state of charge measurement based on the self discharge of
the battery.
6. A method of determining the state of charge of a lead acid
battery comprising: integrating coulombs charging and discharging
the battery to determine a first state of charge measurement;
adjusting the first state of charge measurement using a Peukert
factor; measuring the open circuit voltage to determine a second
state of charge measurement; determining the operation of the
battery and using the first state of charge or second state of
charge measurement as the actual state of charge measurement based
on the operation of the battery.
7. The method of claim 6 further comprising compensating the first
state of charge measurement for temperature effects.
8. The method of claim 6 further comprising compensating the first
state of charge measurement based on charge efficiency variations
caused by regeneration of the lead acid battery.
9. The method of claim 6 further comprising the step of
compensating the first state of charge measurement by comparing the
first state of charge measurement to the second state of charge
measurement.
10. The method of claim 6 further comprising compensating the first
state of charge measurement based on battery capacity
degradation.
11. The method of claim 6 further comprising the step of mapping
the open circuit voltage to the second state of charge.
12. A method of determining the state of charge of a lead acid
battery operating in a vehicle, the method comprising: integrating
current charging and discharging the battery to determine a first
state of charge measurement; adjusting the first state of charge
measurement using a Peukert relationship; measuring the open
circuit voltage to determine a second state of charge measurement
by using a set of data points mapping open circuit voltage to state
of charge for the battery; and determining which of the first state
of charge measurement or the second state of charge measurement
more accurately represents the actual state of charge of the
battery.
13. The method of claim 12 wherein the vehicle is a hybrid electric
vehicle.
14. The method of claim 13 wherein the vehicle is an electric
vehicle.
15. The method of claim 12 further comprising the step of
determining the second state of charge measurement during discharge
conditions for the battery.
16. The method of claim 12 further comprising the step of
compensating the second state of charge measurement based on
voltage shifts after charging of the battery.
17. The method of claim 12 further comprising the step of
compensating the second state of charge measurement based on
battery resistance changes.
Description
TECHNICAL FIELD
[0001] The present invention outlines a method and apparatus to
determine the state of charge (SOC) of a lead-acid battery
operating in a hybrid electric vehicle (HEV).
BACKGROUND OF THE INVENTION
[0002] In today's automotive market, a variety of propulsion or
drive technologies can be used to power vehicles. The technologies
include internal combustion engines (ICEs), electric drive systems
utilizing batteries and/or fuel cells as an energy source, and
hybrid systems utilizing a combination of internal combustion
engines and electric drive systems. Each propulsion system has
specific technological, financial, and performance advantages and
disadvantages, depending on the state of energy prices, energy
infrastructure developments, environmental laws, and government
incentives.
[0003] The increasing demand to improve fuel economy and reduce
emissions in present vehicles has led to the development of
advanced hybrid vehicles. Hybrid electric vehicles (HEV) are
classified as vehicles having at least two separate power sources,
typically an internal combustion engine and an electric traction
motor. A hybrid electric vehicle will generally operate, with a
high-voltage battery pack (.gtoreq.42 V) operating an electric
motor running in conjunction with the ICE.
[0004] Hybrid vehicles, as compared to standard vehicles driven by
an ICE, have improved fuel economy and consequently reduced
emissions. During varying driving conditions hybrid vehicles will
alternate between separate power sources, depending on the most
efficient manner of operation of each power source. For example, a
hybrid vehicle equipped with an ICE and an electric motor will shut
down the ICE during a stopped or idle condition, allowing the
electric motor to restart the ICE and eventually propel the
vehicle, improving fuel economy for the hybrid vehicle.
[0005] Hybrid vehicles are broadly classified into series or
parallel drive trains, depending upon the configuration of the
drive trains. In a series drive train utilizing an ICE and an
electric traction motor, only the electric motor drives the wheels
of a vehicle. The ICE converts a fuel source to mechanical energy
to turn a generator, which converts the mechanical energy to
electrical energy to drive the electric motor. In a parallel hybrid
drive train system, two power sources such as an ICE and an
electric traction motor operate in parallel to propel a vehicle.
Generally, a hybrid vehicle having a parallel drive train combines
the power and range advantages of a conventional ICE with the
efficiency and electrical regeneration capability of an electric
motor to increase fuel economy and lower emissions, as compared
with a traditional ICE vehicle.
[0006] Battery packs having secondary/rechargeable batteries are an
important component of hybrid vehicle systems, as they enable an
electric motor/generator (MoGen) to store braking energy in the
battery pack during regeneration and charging by the ICE. The MoGen
utilizes the stored energy in the battery pack to propel or drive
the vehicle when the ICE is not operating. During operation, the
ICE will be shut on and off intermittently, according to driving
conditions, causing the battery pack to be constantly charged and
discharged by the MoGen. The state of charge (SOC, defined as the
percentage of the full capacity of a battery that is still
available for further discharge) is used to regulate the charging
and discharging of the battery
[0007] Currently, the most cost-effective, commercially-ready
battery for HEV applications is the lead-acid battery. Lead-acid
batteries have been widely used in the automotive industry for
starting-lighting-ignition applications in the last hundred years.
However, in a hybrid application, the power loads and usage are
much heavier than that used in previous lead-acid battery
applications. To operate efficiently in HEV applications, a
lead-acid battery needs to operate near its optimal SOC to maximize
its discharge and charge power capabilities.
[0008] The discharge and charge reactions in the lead-acid battery
are not symmetric, as in other battery technologies such as
nickel-metal hydride and lithium-ion batteries. That is, the
discharge battery resistance is typically lower than the charge
battery resistance, which includes secondary gassing reactions at a
SOC>60%. Consequently, it is difficult to predict the SOC based
on charge voltages. Accordingly, a voltage-based SOC for a
lead-acid battery is usually based on discharge data only. However,
in a HEV application where there are continual regeneration events,
the battery discharge voltages are shifted higher due to increased
concentration of sulfuric acid after a charge. The present
invention provides a method and apparatus to accommodate the effect
of regeneration on the discharge voltage, and thus predict a more
accurate and consistent SOC.
SUMMARY OF THE INVENTION
[0009] The present invention integrates three independent methods
for determining the battery SOC of an electric or HEV during
vehicle operation. The three independent methods include: a
current-based SOC (ISOC) based on amp-hour (Ah) integration; a
voltage-based SOC (VSOC) based on a calculated open circuit voltage
(OCV); and a rest-based SOC (RSOC) based on a measured OCV after
the vehicle has been powered-off. The present invention is
optimized to determine the SOC of a lead-acid battery using factors
uniquely relevant to a lead-acid battery. For example, the present
ISOC method accommodates for charge inefficiency due to secondary
gas reactions and current (Peukert) effect on the discharge
capacity. The VSOC and RSOC calculations in the present invention
consider the discharge data in predicting the OCV at any time
during operation of a battery and compensate for regeneration
effects.
[0010] The SOC determination of the present invention relies
heavily on the VSOC and is adaptive to changes in the battery due
to temperature and aging. The method of the present invention will
periodically reset the ISOC to match the VSOC because of possible
errors in amp-hour integration from small inaccuracies in current
measurements and/or charge inefficiency of the battery over time.
Furthermore, when the vehicle is off, a measured open circuit
voltage can be correlated to the RSOC. The difference in the RSOC
and the VSOC provides the basis for adaptation of battery pack
capacity due to battery degradation resulting from hybrid cycling
at partial SOCs and/or aging.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] FIG. 1 is a diagrammatic drawing of the hybrid electric
vehicle of the present invention;
[0012] FIG. 2 is a flow chart of the current-based SOC method
(ISOC);
[0013] FIG. 3 is a flow chart of the voltage-based SOC method
(VSOC); and
[0014] FIG. 4 is a flow chart of the rest-based SOC method
(RSOC).
DESCRIPTION OF THE PREFERRED EMBODIMENT
[0015] FIG. 1 is a diagrammatic drawing of a hybrid vehicle 14 of
the present invention. The hybrid vehicle 14 includes a battery
pack 16 having a single battery module or individual battery
modules. In the preferred embodiment, the battery pack 16 comprises
a plurality of lead acid battery modules connected in series. In
alternate embodiments of the present invention, the battery pack 16
may comprise any known battery technology including, but not
limited to nickel metal hydride, lithium-ion, and lithium polymer
batteries. An energy management controller (EMC) 15 monitors the
current, voltage, state of charge (SOC), and power output of the
battery pack 16.
[0016] A motor generator (MoGen) 20 is dynamically coupled to an
internal combustion engine (ICE) 22 and functions as either a motor
to propel the vehicle 14 or a generator to charge the battery pack
16, depending on the operating state of the vehicle 14 (i.e.,
braking, accelerating, or operating at a constant speed on a
highway). The MoGen 20 is preferably an AC induction machine, but
may comprise any known electrical motor/generator technology
including, but not limited to DC machines, synchronous machines,
and switched reluctance machines. MoGens 20 in the preferred
embodiment are located on the rear of the vehicle to drive the rear
wheels 17 and on the front of the vehicle to drive the front wheels
18.
[0017] The MoGens 20 are controlled by an electrical control system
comprising a hybrid system controller 23, DC-DC converters 24 and
power inverter modules (PIM) 25. The EMC 15 communicates with the
hybrid system controller 23 and power inverter modules 25 by
various digital and analog signals to provide voltage, current,
and/or power output/input limits for the battery pack 16 based on a
SOC measurement. In alternate embodiments of the present invention,
the EMC 15, hybrid system controller 23, DC-DC converters 24, and
power inverter modules 25 may be configured as a unitary system.
The EMC 15 and hybrid system controller 23 may comprise any type of
control module or vehicle controller known in the art and are
equipped with nonvolatile memory (NVM), random access memory (RAM),
discrete and analog input/output (I/O), a central processing unit,
and communications interfaces for networking within an automotive
communications network.
[0018] In generator mode, the MoGens 20 generate electrical energy
that is transferred to the battery pack 16 and the DC-DC converters
24 by the EMC 15, controller 23 and inverter modules 25. The EMC
15, controller 23 and inverter modules 25 determine the direction
of current flow for the MoGen 20, according to the vehicle 14
operating state. The DC-DC converters 24 provide and regulate the
DC Bus that is pulse-width-modulated by the inverter modules 25 to
supply time varying current to the MoGens 20. In a regeneration
state (such as during braking) or charging condition, current will
flow from the MoGens 20, via the inverter modules 25, to charge the
battery pack 16 and provide current to the DC-DC converters 24. In
a state where the MoGens 20 are needed to provide propulsion,
current will flow from the battery pack 16 to the MoGens 20, via
the DC-DC converters 24 and inverter modules 25, to power the
MoGens 20.
[0019] In the preferred embodiment of the present invention, the
SOC of the battery pack 16 is dynamically tracked to determine when
to charge the battery pack 16. The EMC 15 and hybrid controller 23
of the present invention will control a battery pack's
state-of-charge (SOC) near 50% to 70% so that the charge acceptance
and efficiency during regenerative braking can be realized.
However, controlling the battery pack 16 to any SOC percentage is
within the scope of the present invention.
[0020] As discussed previously, the methods executed by the EMC 15
to determine the SOC of the battery pack 16 includes a
current-based SOC method and measurement (ISOC), a voltage-based
SOC method and measurement (VSOC), and a rest SOC method and
measurement (RSOC).
[0021] Current-Based SOC (ISOC) Method
[0022] FIG. 2 is a flow chart of the ISOC method of the present
invention. The current-based SOC method uses the base equation
defined as: 1 ISOC = ( 1 - Ah_Compensated BP_Capacity _Compensated
) * 100 % [ 1 ]
[0023] to determine a current-based SOC. The term Ah_Compensated is
an accurate amp-hour count of the charge going in and out of the
battery pack 16 starting from a fully-charged state, and the term
BP_Capacity_Compensated is the usable capacity of the battery pack
16. Ah_Compensated is bounded between zero at 100% SOC and
BP_Capacity_Compensated at 0% SOC. The sign convention in the
present invention is written as negative for discharge and positive
for charge. Accordingly, the sign of both Ah_Compensated and
BP_Capacity_Compensated is negative. They are compensated for
various factors such as current, temperature, charge efficiency,
parasitic losses, and self-discharge, as will be described later in
the specification.
[0024] Because the capacity of a lead-acid battery is dependent on
the discharge current known as the Peukert effect, it is pertinent
to determine the capacity at various discharge rates. The battery
pack 16 capacity will change depending on the most recent average
discharge current, such as a discharge current measured in the last
five minutes. For example, if the battery pack 16 had been
initially discharging at a fast rate and then switched to a slower
rate, more capacity would be available at the slower rate. Even a
rest period can increase the battery pack's 16 discharge capacity.
Starting at block 102 in FIG. 2, an average discharge current may
be determined from primarily the discharge portion of the drive
cycle. The average discharge current should be representative of
the rate at which the battery pack 16 capacity is being removed.
Therefore, the time period over which the cumulative discharge is
summed should be sufficient to reflect the recent discharge history
of the battery pack 16 such as approximately five minutes. Also,
the time constants for updating the average discharge current
should be long enough such that the average discharge current is
not constantly changing. The formula used in block 102 to calculate
the average discharge current is: 2 Discharge_Current _Average = i
( Discharge_Ah ) i i ( Discharge_time ) i [ 2 ]
[0025] where Discharge_Current Average is the average discharge
current for the battery, Discharge_Ah is the integrated amp-hour
using the measured discharge currents, and Discharge_time is the
time period over which the Discharge_Ah is measured.
[0026] For approximately the first five minutes of discharge, the
initial average discharge current should be set to the average
discharge current from the previous driving cycle. If the average
discharge current from the previous driving cycle is not available,
the default value can be set to the C-rate, which is defined as the
discharge current needed to deplete the capacity of the battery
back 16 in approximately one hour.
[0027] Typically, the battery capacity discharged at the C-rate is
considered the usable capacity in a HEV application and is defined
as BP_Capacity_Default at a reference temperature such as
25.degree. Celsius. To compensate for the effect of discharge
current and temperature, a Peukert factor or Peukert_Factor is
introduced in block 102 and a temperature compensation factor for
the battery capacity or Capacity_Temp_Factor is introduced in block
104 by using look-up tables or formulas. The
BP_Capacity_Compensated calculation in block 106 uses the following
formula:
BP_Capacity_Compensated=BP_Capacity_Default*Peukert_Factor*Capacity_Temp_F-
actor [3]
[0028] The Peukert_Factor is based on the Peukert Equation:
(I.sub.D,avg).sup.n*t=C [4]
[0029] where I.sub.D,avg is the Discharge_Current_Average, n is the
Peukert slope, t is the total discharge time, and C is a constant.
For a lead-acid battery, the Peukert slope n is typically between
1.1 and 1.3 at 25.degree. C. and is a function of temperature. The
Peukert slope usually increases with decreasing temperatures. Thus,
the Peukert relationship needs to be characterized for the selected
battery in the temperature range of vehicle operation.
[0030] If the BP_Capacity_Default is based on a C-rate discharge
(i.e., I.sub.D,avg=C-rate), then the Peukert_Factor can be
calculated for a different Discharge_Current_Average,
I.sub.D,avg,2, by
Peukert_Factor=(I.sub.D,avg,1/I.sub.D,avg,2 or
Peukert_Factor=(C-rate/Discharge_Current_Average).sup.n-1 [5]
[0031] The Peukert relationship characterized at different
temperatures can also provide the Capacity_Temp_Factor, which takes
into account the effect of temperature on the battery capacity.
Using the C-rate capacities at different temperatures,
Capacity_Temp_Factor (T)=Battery_Capacity (T)/BP_Capacity_Default
(25.degree. C.) [6]
[0032] Block 108 then adjusts the Capacity_Weight_Factor, which
will be discussed in conjunction with FIG. 4.
[0033] Besides having an accurate prediction of the battery pack 16
capacity, it is important to have an accurate amp-hour count of the
charge going in and out of the battery pack 16 to determine an
accurate ISOC. Numerous sources of uncertainty such as charge
inefficiency due to secondary gassing reactions, parasitic losses,
and self-discharge exist. Errors generated by these sources of
uncertainty may be mitigated by using various methods of
compensation.
[0034] Referring to FIG. 2 at block 110, the uncorrected amp-hour
counts are added separately for discharge and charge
(regeneration). The following terms are calculated for discharge
currents, I<0:
Ah_Discharge.sub.i=(I.sub.i+I.sub.i-I)/2*(t.sub.i-t.sub.i-I)
[7]
[0035] and for charge currents, I>0:
Ah_Regen.sub.i=(I.sub.i+I.sub.i-I)/2*(t.sub.i-t.sub.i-I) [8]
[0036] where Ah_Discharge is the amp-hour count for a battery pack
16 from discharge and Ah_Regen is the amp-hour count for a battery
pack 16 from charge or regeneration.
[0037] Because of possible secondary gassing reactions during
regeneration, particularly at a SOC>60%, not all charge current
will go towards active material conversion. A charge efficiency
factor, Charge_EfficiencyFactor, is introduced at block 112 as a
function of the battery pack 16 SOC, Pack_SOC, and regeneration
current, Regen_Current. The Charge_Efficiency_Factor is generally
lower at high Pack_SOC and high Regen_Current. Accordingly, the
compensated amp-hour counts during regeneration,
Ah_Regen_Compensated.sub.i, is calculated by the following
formula:
Ah_Regen_Compensated.sub.i=Charge_Efficiency_Factor*Ah_Regen.sub.I
[9]
[0038] At block 114, the cumulative amp-hours, Ah_Cumulative, are
totaled by the following formula:
Ah_Cumulative=.SIGMA.(Ah_Discharge.sub.i+Ah_Regen_Compensated.sub.i)
[10]
[0039] Note: Because of the sign convention, Ah_Discharge.sub.i is
negative and Ah_Regen_Compensated, is positive.
[0040] During a prolonged rest, the battery pack 16 loses
additional capacity from parasitic drains due to electronic
equipment on-board the vehicle and from self-discharge. These
losses are considered in the amp-hour count in blocks 116 and 118.
The amp-hour parasitic losses, Ah_Parasitic_Loss, are calculated by
the following formula at block 116:
Ah_Parasitic_Loss=Parasitic_Rate*Rest_Time [11]
[0041] where Parasitic_Rate is negative and needs to be
pre-determined for the specific vehicle electrical loads during off
time, and Rest_Time is the total time when the vehicle is not in
operation. The self-discharge rate of the battery,
Self_Discharge_Rate, is dependent on the battery's SOC and
temperature and needs to be pre-determined for the selected
battery. The amp-hour self-discharge losses, Ah_Self_Discharge, are
calculated by the following formula in block 118:
Ah_Self_Discharge=Self_Discharge_Rate*Rest_Time [12]
[0042] Note: The amp-hours associated with parasitic drains and
self-discharge are negative because they are depleting the
battery.
[0043] After compensating for charge efficiency, parasitic and
self-discharge losses, the compensated amp-hour count,
Ah_Compensated, in block 120 may be calculated by the formula:
Ah_Compensated=Ah_Cumulative+Ah_Parasitic_Loss+Ah_Self Discharge
[13]
[0044] Because of small errors in the amp-hour count, parasitic
loss and self-discharge calculations building up over time as well
as changes in the battery pack 16 capacity occurring with age,
periodic adjustments of both the numerator and denominator in
equation [1] are needed. The amp-hour correction, Ah_Correction, at
block 122 and Capacity_Weight_Factor in block 108 accommodate these
errors based on differences in the SOCs determined by the three
methods. Blocks 122 and 108 will be further described in
conjunction with FIGS. 3 and 4, respectively. Thus, at block 124,
the ISOC for the battery pack 16 may be determined by the formula:
3 ISOC = ( 1 - Ah_Compensated + Ah_Correction BP_Capacity
_Compensated * Capacity_Weight _Factor ) * 100 % [ 14 ]
[0045] where the default Ah_Correction is initially zero and
Capacity_Weight_Factor is initially one. The Ah_Correction is
adjusted based on the difference between the VSOC and the ISOC
while the Capacity_Weight_Factor is based on the difference between
the VSOC and the RSOC. These adjustments are discussed in the VSOC
and the RSOC portions of the present specification. Large
fluctuation in these factors can be avoided by limiting the
allowable rate of change for these factors.
[0046] Voltage-Based SOC (VSOC) Method During Discharge
[0047] It is well known that the open circuit voltage (OCV) of a
lead-acid battery is dependent on the sulfuric acid concentration
at the electrode surfaces, which correlates to a lead-acid
battery's SOC. However, the SOC versus OCV correlation is dependent
on several factors such as the prior current and time at which the
OCV is measured. In an HEV application, there are continual
regeneration events, which can further complicate the correlation.
The voltage-based SOC method of the present invention is shown in a
flow chart in FIG. 3.
[0048] The VSOC calculation is deactivated under certain conditions
as shown at the start of the VSOC flow chart at block 202. First,
the VSOC measurement for the present invention is determined using
discharge data only. For charge (i.e., I>0), the present
invention at block 204 will rely on the ISOC calculation to
determine the SOC of the battery pack 16. During charge, the module
resistance includes an added complexity associated with the
secondary gassing reactions occurring at the electrodes. The
gassing polarizations can be substantially larger than the
activation polarization particularly at a SOC>60%; therefore,
the gassing polarizations are difficult to account for in the OCV
back-calculation. Accordingly, the voltage-based SOC uses the
discharge and not regeneration data in the VSOC determination. The
SOC determination of the battery pack 16 should rely on ISOC during
regeneration events. Low discharge currents may also back-calculate
to an abnormally high OCV particularly after a regeneration event
because of their low polarization voltages. Therefore, for low
discharge currents whose magnitude is less than the C-rate, the
VSOC will not be updated.
[0049] The VSOC of the present invention is preferably not
determined for the battery pack 16 after long rests when the
sulfuric acid at the electrode surface has equilibrated with the
bulk electrolyte. The additional polarization associated with the
diffusion of the sulfuric acid back to the electrode surface will
give a false, lower SOC reading. The VSOC of the present invention
is preferably deactivated initially until the conditions at the
electrode surfaces stabilize, which can be until the magnitude of
the Ah_Discharge is greater than the BP_Capacity as also indicated
in block 202. The SOC determination of the battery pack 16 in the
present invention should rely on the ISOC when the VSOC is
deactivated.
[0050] The SOC of the worst performing battery module in the
battery pack 16 will dictate the SOC of the battery pack 16.
Referring to FIG. 3, the weakest battery module of the battery pack
16 is determined at block 206. Accordingly, the lowest module's
voltage and corresponding current are used to determine the VSOC.
The lowest module is selected by comparing each module voltage with
a calculated normalized module voltage. The normalized module
voltage uses values of pack voltage and pack current that are
synchronized with the module voltages being evaluated. The total
strap resistances, Total_Strap_Resistance, between the modules in
the battery pack are also measured. The normalized module voltage,
Module_Voltage_Normalized, and the difference in module voltage,
Module_Voltage_Delta, for each module are calculated by the
following equations:
Module_Voltage_Normalized=((Pack_Voltage-(Total_Strap_Resistance*Current))-
/Module_Number [15]
Module_Voltage_Delta=(Module_Voltage_Normalized-Module_Voltage)
[16]
[0051] where Module_Number is the total number of modules in
battery pack 16 and Module_Voltage is the voltage for each separate
battery module. The battery module with the highest
Module_Voltage_Delta is selected as the lowest module. This module
is used to determine the VSOC. The current-voltage data for this
module are synchronous.
[0052] To correlate to the VSOC, it is preferable to use a measured
OCV after a discharge pulse greater than C-rate in magnitude. When
available during vehicle operation, the instantaneous OCV is
recorded (such as measured 0.2 second after the current load is
removed) and the method of the present invention proceeds to block
212 to factor in the OCV transient behavior.
[0053] However, during most vehicle operation, the battery pack 16
is usually under a load, so it is difficult to measure the OCV
directly. Fortunately, the OCV can be calculated if the battery
module resistance is known:
V-OCV=I*Module_Resistance [17]
[0054] Since the resistance of each battery module,
Module_Resistance, is dependent on SOC, module temperature, and
module age, the battery module resistance is continually updated
during vehicle operation by dividing delta V by delta I at short
time periods (.ltoreq.1 second, preferably 0.2 second) at block 208
using the equation: 4 Module_Resistance i = V i - V i - 1 I i - I i
- 1 [ 18 ]
[0055] The Module_Resistance is an overall battery resistance
associated with the electrochemical reactions occurring at the
positive and negative plates including ohmic, crystallization,
activation (kinetics), and concentration (diffusion) resistances.
Equation 18 assumes that the Module_Resistance.sub.i and OCV have
not changed significantly in the time interval, t.sub.i-t.sub.i.
This is generally true for very short periods such as 0.2 second.
It is important that a consistent time interval be used to
calculate the Module_Resistance.sub.i because the battery voltage
changes continually with time.
[0056] Furthermore, the Module_Resistance.sub.i is calculated for
discharge currents whose magnitude is greater than the C-rate. The
module resistance is not updated if the current difference is very
small (e.g., delta I<C/5), low discharge currents, and charge
currents. In these situations, the previous module resistance value
should be used. To prevent large fluctuations, the resultant
Module_Resistance.sub.i value may be filtered, changed gradually,
and bounded between its 100% SOC impedance @1000 Hz and five times
that initial value.
[0057] At block 210, the open circuit voltage, OCV Instant, is
calculated from the Module_Resistance.sub.i using the following
equation:
OCV_Instant=V.sub.i-(I.sub.i*Module_Resistance.sub.i) [19]
[0058] For a time interval of 0.2 second when the
Module_Resistance.sub.i is measured, the OCV_Instant versus SOC
correlation is dependent on the magnitude of the prior discharge
current. If the OCV_Instant is allowed to relax for a longer time
period, such as for one minute, the SOC-OCV correlation is less
dependent on the prior discharge current. The OCV transient
behavior may be characterized for a time period of a minute or
longer after different discharge currents and SOC.
[0059] When a current load is removed, the OCV value changes
logarithmically with time,
OCV(t)=OCV_Transient_Slope*ln(t)+OCV.sub.--ls [20]
[0060] where t is time in seconds and OCV_Transient_Slope and
OCV_ls are fitted parameters of the measured OCV plotted against
rest time. This transient behavior is associated with the diffusion
of the sulfuric acid between the electrode surface and bulk
electrolyte. The OCV_Transient_Slope indicates the direction and
rate at which the OCV relaxes to its semi-steady state value. The
correlation coefficients from these curve fits are usually greater
than 0.98. Following a discharge pulse, the OCV relaxes to a higher
value, so the transient slope is positive. Whereas following a
charge pulse, the OCV relaxes to lower value, and the transient
slope is negative. Block 212 contains a look-up table of the
pre-determined OCV_Transient_Slope values as functions of both the
prior discharge current and SOC.
[0061] Equation 20 allows the OCV to be calculated for any time,
t.sub.2, once an instantaneous OCV is measured or calculated at
t.sub.1,
OCV(t.sub.2)=OCV(t.sub.1)+OCV_Transient_Slope*ln(t.sub.2/t.sub.1)
[21]
[0062] where for instance, t.sub.1=0.2 s when OCV_Instant is
calculated or measured and t.sub.2=60 s when the SOC versus OCV
correlation is calibrated.
[0063] The OCV_Transient_Slope is also a function of temperature
and can be temperature-compensated by the
OCV_Transient_Slope_Temp_Factor. Accordingly, the compensated open
circuit voltage transient slope may be determined by the formula: 5
OCV_Transient _Slope _Compensated = OCV_Transient _Slope _Temp
_Factor * OCV_Transient _Slope _ 25 C [ 22 ]
[0064] Once the OCV_Instant is calculated from equation [19] and
the OCV_Transient_Slope_Compensated is determined, the OCV is
calculated for the time period (such as 60 seconds) used to
calibrate the SOC versus OCV correlation at block 214 using the
equation:
OCV.sub.--60
s=OCV_Instant+OCV_Transient_Slope_Compensated*ln(60/Instant_t- ime)
[23]
[0065] In the preferred embodiment of the present invention,
Instant_time is 0.2 second and calibration time is 60 seconds.
[0066] At block 216, the OCV.sub.--60 s is compensated for the
voltage shift caused by charge or regeneration events. During
charge, sulfuric acid is being produced at the electrode;
therefore, the local concentration of sulfuric acid is high causing
the voltage to be high when discharge is initialized. As the
sulfuric acid is being consumed during the discharge, the voltage
settles back down to a point when it can be calibrated to the SOC
(i.e., when -Ah_Discharge/Ah_Regen.gtoreq.1)- . This ratio
Ah_Discharge/Ah_Regen is updated every time charge occurs.
[0067] For -Ah_Discharge/Ah_Regen<1, the OCV.sub.--60 s is
compensated for the regeneration shift by the formula: 6 OCV_ 60 s
_Shifted = OCV_ 60 s + ( 1 - Ah_Discharge Ah_Regen ) * Regen_Shift
[ 24 ]
[0068] where Ah_Discharge and Ah_Regen are the coulomb counts of
the present discharge and prior regeneration pulses, and
Regen_Shift is negative to shift the OCV lower. Note that there is
no Regen_Shift correction when -Ah_Discharge/Ah_Regen.gtoreq.1.
[0069] The Regen_Shift has been found to be dependent on the
magnitude of the prior regeneration current and SOC. With the
exception of low and high SOCs, the Regen_Shift is fairly constant
between 20% and 80% SOC range, where most hybrid battery operation
occurs.
[0070] If the OCV.sub.--60 s_Shifted fluctuates significantly over
a short period of time (such as one to five minutes), a smoothing
technique such as averaging may be used to get an OCV.sub.--60
s_Shifted Average.
[0071] The correlation of the OCV to SOC can be pre-determined from
OCV measurements following a range of discharge pulses at different
SOC. To maintain SOC during these measurements, each discharge/rest
pulse should be followed by an equivalent charge/rest pulse. For
example, at each SOC, voltage measurements can be made for pulses
of 10-second discharge/60 second rest/10 second charge/60 second
rest for currents ranging from C-rate to 10 C-rate. The 60-second
OCV following the discharge pulses would be correlated to the SOC,
thus generating a SOC-OCV calibration curve, which can be described
as a fitted equation or in a look-up table. A set of SOC-OCV
calibration curves shall be pre-determined for a variety of
conditions including, but not limited to temperature and battery
aging.
[0072] Because of the voltage hysteresis effect, the SOC versus OCV
relationship is dependent on whether the battery pack 16 is
operating in a charge decreasing or charge increasing mode. To
assess the degree of voltage hysteresis, the SOC-OCV calibration
curves are determined separately for a charge decreasing and a
charge increasing operation at different temperatures. The voltage
difference between the charge decreasing and charge increasing
SOC-OCV calibration curves is fairly constant, so it may be more
convenient to use one SOC-OCV calibration curve for each
temperature rather than both. Therefore, if the charge decreasing
SOC-OCV calibration curve is selected, then an offset can be used
to shift any charge increasing data to the charge decreasing
calibration curve.
[0073] A Hybrid_Current_Average is defined in block 218 to
determine the mode of battery operation. It can be calculated by
dividing the cumulative (discharge and charge) amp-hour by time and
updated by a time interval such as every five minutes. If the
Hybrid_Current_Average is zero, the battery pack 16 is in a
charge-sustaining mode of operation. If the Hybrid_Current_Average
is less than zero, the battery pack 16 is in a charge decreasing
mode, and if the Hybrid_Current_Average is greater than zero, the
battery pack 16 is in a charge increasing mode.
[0074] Accordingly at block 220, for a charge increasing operation
(i.e., Hybrid_Current_Average>0), a charge increasing offset,
Charge_Increasing_Offset, may be introduced to shift the
OCV.sub.--6 s_Shifted_Average from the charge increasing
calibration curve to the charge decreasing one using the
equation:
OCV 60 s_Compensated=OCV.sub.--60
s_Shifted_Average+Charge_Increasing_Offs- et [25]
[0075] The Charge_Increasing_Offset is zero for charge decreasing
or charge sustaining data (i.e., Hybrid_Current_Average.ltoreq.0).
OCV.sub.--60 s_Compensated is correlated to VSOC using the charge
decreasing SOC-OCV calibration curve, which is a function of
temperature and battery aging, in block 222.
[0076] The VSOC and the ISOC are compared at block 224. If the VSOC
is similar to the ISOC, the battery pack SOC or Pack_SOC is set
equal to the VSOC to be used as the correct SOC for the battery
back 16 in block 226. If the VSOC and the ISOC are not similar,
then there will be a correction using Ah_Correction for the ISOC
value at block 228.
[0077] The Ah_Correction is based on the difference between the
VSOC and the ISOC. The initial or default Ah_Correction=0 and is
adjusted by
Ah_Correction=(ISOC-VSOC)*BP_Capacity_Compensated [26]
[0078] Rest-Based SOC (RSOC) Method Based on Measured OCV
[0079] The SOC can be adjusted when there is an opportunity to
measure a true OCV. FIG. 4 is a flow chart of the Rest-based SOC
method of the present invention. At block 302, if the vehicle 14 is
on, then the ISOC and the VSOC will be determined at block 304 as
previously described, and the RSOC method will not be executed. If
the vehicle 14 is off, then at the end of vehicle operation in
block 306, the control module shall record the OCV of the lowest
module at 60 seconds, OCV.sub.--60 s_Rest, after the last current
pulse (preferably discharge). A large discharge (>C-rate) pulse
will give a relatively better SOC correlation than a charge
current. Similar to the VSOC method above, the OCV.sub.--60 s_Rest
may need to be shifted based on the latest ratio
-Ah_Discharge/Ah_Regen (<one). At block 308 the OCV rest value,
OCV.sub.--60_Rest_Shifted, is shifted using the following equation:
7 OCV_ 60 s _Rest _Shifted = OCV_ 60 s _Rest + ( 1 - Ah_Discharge
Ah_Regen ) * Regen_Shift [ 27 ]
[0080] No Regen_Shift is necessary if
-Ah_Discharge/Ah_Regen.gtoreq.one.
[0081] Using the prior Hybrid_Current_Average to determine if the
battery pack 16 is charge decreasing or charge increasing at block
310, the OCV.sub.--60 s_Rest_Compensated is determined at block 312
based on the equation:
OCV.sub.--60 s_Rest_Compensated=OCV.sub.--60
s_Rest_Shifted+Charge_Increas- ing_Offset [28]
[0082] where Charge Increasing Offset is zero for charge decreasing
or charge sustaining operation. At block 314, the RSOC is
determined by using the OCV.sub.--60 s_Rest_Compensated based on
the SOC-OCV calibration curve depending on the temperature and
battery age.
[0083] The RSOC is compared with the VSOC at block 316. If they are
similar, then the battery pack 16 SOC is set to the RSOC at block
318. However, if they are dissimilar, then the
Capacity_Weight_Factor (Range=0.2-1.2) is determined based on the
difference between the VSOC and the RSOC. The initial or default
Capacity_Weight_Factor=1.0 and is adjusted in block 320 in FIG. 4
by 8 Capacity_Weight _Factor = ( 1 - VSOC 1 - RSOC ) [ 29 ]
[0084] During vehicle operation, the SOC algorithm relies heavily
on the VSOC as the Pack_SOC. In the ISOC calculation, there are two
sources of error in the Ah_Compensated and BP_Capacity_Compensated
values. The SOC strategy incorporates an Ah_Correction and/or
Capacity_Weight_Factor adjustments to accommodate for differences
in the SOCs determined by the three methods. However, under certain
circumstances, VSOC is not valid.
[0085] To summarize, for low discharge currents, all charge
currents, and after prolong rests:
Pack.sub.--SOC=ISOC [30]
[0086] For large discharge currents greater than the C-rate:
Pack.sub.--SOC=VSOC [31]
[0087] When there is a large discharge pulse prior to the end of
vehicle operation:
Pack SOC=RSOC [32]
[0088] While this invention has been described in terms of some
specific embodiments, it will be appreciated that other forms can
readily be adapted by one skilled in the art. Accordingly, the
scope of this invention is to be considered limited only by the
following claims.
* * * * *