WO2014018048A1

WO2014018048A1 - Battery management system

Info

Publication number: WO2014018048A1
Application number: PCT/US2012/048503
Authority: WO
Inventors: Monika Alicia Alexandria MINARCIN; Brian Conway
Original assignee: International Engine Intellectual Property Company, Llc
Priority date: 2012-07-27
Filing date: 2012-07-27
Publication date: 2014-01-30

Abstract

A battery management system and method for use with one or more batteries are described. The system can include, for example, a processor and sensors. The sensors are coupled to the processor and to the one or more batteries. The processor causes the one or more batteries to be drained over a period of time. The sensors take one or more voltage measurements of the one or more batteries during the period of time. Then, the processor determines an effective resistance of the one or more batteries based on the one or more voltage measurements and determines a battery state-of-health of the one or more batteries based on the determined effective resistance.

Description

D7220

BATTERY MANAGEMENT SYSTEM

BACKGROUND

[0001] Battery State-of-Health (SoH) is typically an undetermined indicator that can be used to describe the health of a battery. It is an indication of a current state of the battery compared to a new state of the battery.

[0002] Most attempts to represent SoH use a capacity calculation or add effective ohmic and charge transfer as calculated from an algorithm such as the 6-parameter circuit model, which is a type of equivalent circuit model. Such a model was limited in application to and focused on a simulation of a single electrochemical cell. However, applying the 6-parameter circuit model would be challenging, if not impractical, if applied to a simulation of a multiple electrochemical cell system.

[0003] The 6-parameter equivalent circuit model is greatly simplified for purposes of computational expediency. Better fidelity in predicting electrochemical cell behavior can be achieved through the use of additional resistance-capacitance (RC) circuits in the series string of components that comprise the model. However, this is only an estimation and extra factors, including variable, real-time and/or unpredictable factors, need to be considered. A balance must be struck between the computational resources the model consumes and its ability to emulate the behavior of the terminal voltage particular in a real-time vehicle environment.

[0004] For many batteries, which contain multiple electrochemical cells, modules, or packs, being considered for hybrid vehicle applications, at least three separate time constants can be identified with the transient response of the battery, which would dictate the use of three RC elements in the model which is labor intensive to calibrate and to develop. Furthermore, any changes in the battery's electrochemical cell construction or tolerances might render a particular version or model of an algorithm useless. Typically, two of the RC elements with the longest time constants are used and the remaining RC element is ignored for the model. As a consequence of its short time constant, the resistive contribution of this element to the battery D7220 terminal voltage usually becomes lumped together with Ohmic resistance (e.g., Ro or R_0hm) and the charge transfer resistance value appears much higher than would normally be predicted.

[0005] Such a model would be inapplicable to medium duty and heavy duty applications, for example, which can see low temperatures, high power and high current RMS and would be inapplicable to vehicles with a large number of electrochemical cells. In addition, the 6- parameter equivalent circuit model cannot adequately address individual driving factors including, for example, whether the driver is aggressive or conservative, the number of hours of use, the number of hours of rest, etc. The 6-parameter equivalent circuit model would have to predict and be correct, in advance, with respect to all the possible use cycles to predict an SoH value. This is impractical under the 6-parameter equivalent circuit model. The model and the battery allow for different time constants depending on the sign of current (e.g., charge or discharge current), a characteristic which is not attributed to the Ohmic resistance. If different values are required, as has been demonstrated through vehicle testing and energy storage system aging and temperature testing, the modeled charge transfer and Ohmic resistance cannot provide them. In general, the 6-parameter equivalent circuit model cannot be adapted for medium duty and heavy duty truck services because there are a wide variety of vocations and duty cycles that cannot be adequately modeled due, in part, because of the massive number of tests that would have to be run and the prohibitive costs.

[0006] Other attempted models provide that the electrochemical cell be fully charged, but the test is considered destructive: the IEC 61960:2003 provides for an AC method that measures

VA while applying an AC IA, thereby calculating AC resistance; and the IEC 61960:2003 provides for a DC method in which the battery is discharged twice in sequence at different currents and the battery voltage is measure at the end of each discharge, thereby allowing a calculation of a DC resistance given by a ratio of a difference of the two voltage measurements at the end of the discharging and the two currents. D7220

SUMMARY

[0007] Some embodiments described herein relate to systems and methods for determining battery state-of-health (SoH).

[0008] In one embodiment, a battery management system is used with one or more batteries. The system includes, for example, a processor and sensors. The sensors are coupled to the processor and to the one or more batteries. The processor causes the one or more batteries to be drained over a period of time. The sensors take one or more voltage measurements of the one or more batteries during the period of time. Then, the processor determines an effective resistance of the one or more batteries based on the one or more voltage measurements and determines a battery state-of-health of the one or more batteries based on the determined effective resistance.

[0009] Another embodiment provides a method of managing one or more batteries. The method can include, for example, one or more of the following steps: causing one or more batteries to be drained over a period of time; taking one or more voltage measurements of the one or more batteries during the period of time; determining an effective resistance of the one or more batteries based on the one or more voltage measurements; and determining a battery state- of-health of the one or more batteries based on the determine effective resistance.

[0010] In another embodiment, a battery management system is used with one or more batteries. The system includes, for example, means for causing one or more batteries to be drained over a period of time; means for taking one or more voltage measurements of the one or more batteries during the period of time; means for determining an effective resistance of the one or more batteries based on the one or more voltage measurements; and means for determining a battery state-of-health of the one or more batteries based on the determined effective resistance.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 is a block diagram of an embodiment of a battery pack and control systems of a vehicle. D7220

[0012] FIG. 2 is a block diagram of an embodiment of a battery management system illustrated in FIG. 1.

[0013] FIG. 3 is a block diagram of an embodiment of a power control system illustrated in FIG. 1.

[0014] FIG. 4 is a block diagram of an embodiment of a vehicle network.

[0015] FIG. 5 is a flow chart of an embodiment of a process for determining a battery

State-of-Health (SoH).

[0016] FIG. 6 is a flow chart of an embodiment of a process for determining a battery

SoH.

DETAILED DESCRIPTION

[0017] A battery management system is typically implemented into battery operation to optimize battery performance and to mitigate safety hazards.

[0018] In some embodiments, the battery management system monitors characteristics such as pressure, temperature, voltage or current for preset values to protect the battery cell (e.g., an electrochemical cell, unit or element) from operation outside the limits recommended by manufacturer. The battery management system also measures individual battery characteristics to determine information relating to present conditions. The determined information can be, for example, an indication of battery State-of-Health (SoH) state or status. The battery SoH state or status can then be the basis for more advanced tasks such as, for example, displaying battery capacity or battery charge status in a fuel gauge (e.g., an energy system fuel gauge).

[0019] Referring to FIGS. 1-3, a battery pack 100 (e.g., an electrochemical pack) can include, for example, one or more battery modules 110 (e.g., electrochemical modules) and a battery management system 120. Although referred to as a battery pack 100, the battery pack 100 can be, for example, a vehicle energy storage system according to some embodiments. Furthermore, although referred to as a battery management system 120, the battery management D7220 system 120 can be, for example, a vehicle energy storage management system. Each battery module 110 can include, for example, one or more battery cells 130 (e.g., electrochemical cells, units, or elements). The battery management system 120 is coupled to power and/or data inputs and outputs of the battery pack 100. In addition, the battery management system 120 can be coupled to the battery modules 110 and/or to the battery cells 130. For example, the battery management system 120 can be coupled to individual battery cells 130 or stacks of battery cells 130 (e.g., stacks of electrochemical cells, units or elements).

[0020] The battery pack 100 and the battery management system 120 are coupled to the control systems 140 (e.g., control systems of a vehicle). The control systems 130 can include, for example, an engine control system 150, a power control system 160, and/or other control systems.

[0021] In some embodiments, the battery pack 100, the battery management system 120, and the control systems 140 are part vehicles the can include one or more of the following: hybrid electric vehicles (HEVs), electric vehicles (EVs), plug-in electric vehicles (PHEVs), and range-extended electric vehicles (REEVs). The battery cells 130, the battery modules 110 and the battery pack 100 can include rechargeable batteries such as lithium batteries, for example.

[0022] Referring to FIG. 2, the battery management system 120 can include, for example, one or more of the following: one or more processors 170, one or more memories 180, one or more temperature sensors 190, one or more voltage sensors 200, one or more current sensors 210, one or more pressure sensors 220, and other battery management circuitry. The sensors

190, 200, 210, and 220 can be coupled to individual battery cells 130, stacks of battery cells 130, battery modules 110, and power inputs/outputs of the battery pack 100.

[0023] The battery management system 120 can be inside the battery pack 100 and/or can be outside the battery pack 100. The battery management system 120 can be distributed, for example, throughout the battery pack 100. For example, subsystems of the battery management system 120 can be distributed to the battery modules. D7220

[0024] Referring to FIG. 3, the power control system 160 can include, for example, one or more processors 230, one or more memories 240, a converter 250, an inverter 260, a generator 270, a plug-in charger 280, and other power circuitry.

[0025] Referring to FIG. 4, a network configuration is illustrated in which one or more the following are linked by a network 290 (e.g., a vehicular data and/or power network): the processor 170, the memory 180, the battery cells 130, the battery modules 110, the battery pack 100, the control systems 140, and the sensors 190, 200, 210, and 220.

[0026] The processor 170 and 230 can include, for example, one or more of the following: a general processor, a central processing unit, a digital filter, a microprocessor, a digital processor, a digital signal processor, a signal processor, a microcontroller, a programmable array logic device, a complex programmable logic device, a field-programmable gate array, an application specific integrated circuit, and an internal cache or memory. Code, instructions, software, firmware and/or data can be stored in the processor 170 and 230 and/or the memory 180 and 240. [0027] The memory 180 and 240 can include, for example, one or more of the following: a non-transitory memory, a non-transitory processor readable medium, a non-transitory computer readable medium, a read only memory (ROM), a programmable ROM, a random access memory (RAM), a cache, a semiconductor memory, a flash drive, a magnetic memory, an optical memory, an electromagnetic memory, etc. The memory 180 and 240 can be configured to store code, instructions, software, firmware and/or data for use by the processor 170 and 230 and can be external and/or internal to the processor 70 and 230.

[0028] Some of the code, instructions, software, firmware and/or data can be hardwired

(e.g., hardware implementations, hardwired into registers, etc.) and/or can be programmable.

[0029] In operation, the battery management system 120 can monitor the temperature, voltage, current and/or pressure, via the sensors 190, 200, 210, and 220, of the battery cells 130, the battery modules 110, and the battery pack 110. In some embodiments, instead of or in D7220 addition to the sensors 190, 200, 210 and 220, algorithm, emulation, simulation or test results or signals can be input to the battery management system 120.

[0030] The processor 170 of the battery management system 120 can execute code, instructions, software, and/or firmware that is stored in the memory 180, for example, to control the power control system 160, the battery cells 130, the battery modules 110, and/or the battery pack 100 and to measure and/or calculate one or more parameters (e.g., an electrochemical-based parameter) that can be used to determine the SoH state of individual battery cells 130, subsets of battery cells 130, individual battery modules 110, subsets of battery modules 110, and/or the battery pack 100. The determined SoH state can be used to determine whether the one or more battery cells 130, the one or more battery modules 110, and/or the battery pack 100 is due to be replaced or is due for maintenance. The SoH state and/or the recommendation to replace or to maintain the battery cells 130, the battery modules 110, and/or the battery pack 100 can be displayed on a display of the vehicle or on an external display (e.g., a display on diagnostic equipment). [0031] In some embodiments, battery State-of-Health (SoH) status or state is a quantitative indicator or figure of merit used to describe the health of a battery (e.g., a rechargeable battery, an electrochemical cell, module or pack, etc.) In some embodiments, it is an indication of a current state of the electrochemical cell, module, or pack relative to a new state (e.g., unused or new state) of the electrochemical cell, module, or pack. SoH status or state can take into account such factors as, for example, charge acceptance, internal resistance, voltage and/or self-discharge.

[0032] Original equipment manufacturers (OEMs) and/or customers can use SoH status or state to determine whether a battery should be replaced or is due for maintenance. The SoH status or state can be used to meet on-board diagnostics (OBD) requirements including, for example, OBD II requirements. The SoH status or state can be used to determine how much power or energy an electrochemical cell, module, or pack can provide or absorb, for example. D7220

[0033] Referring to FIG. 5, an embodiment of a method for determining the SoH state is described. In step 300, the processor 170 of the battery management system 120 causes the power control system 160 to partially drain one or more battery cells 130, one or more battery modules 110, and/or the battery pack 110. The partial drain is accomplished in a controlled manner, for example, via a constant current over a period of time (e.g., a preset or programmed period of time) at a low rate. The constant current can be followed, for example, by a constant voltage charge.

[0034] In step 310, the processor 170 can measure one or more voltages over the period of time or at a particular time, via the voltage sensors 200, over one or more battery cells 130, one or more modules 110, and/or the battery pack 100.

[0035] In step 320, the processor 170 can then determine an effective discharge resistance of the one or more battery cells 130, the one or more modules 110, and/or the battery pack 100. The effective discharge resistance can be calculated as a change in voltage divided by the current, or as a final voltage, after the period of time, divided by the current. In some embodiments, the effective discharge resistance is not the "accepted" electrochemical or physical discharge resistance, but is a combination of these.

[0036] In some embodiments, the effective discharge resistance is an electro-chemical- based parameter derived from direct measurement or from an algorithm. The effective discharge resistance change substantially with age and thus can be used as a basis for providing an indication of the SoH of a battery cell, module or pack. Changes in the effective discharge resistance can be quantified as indicators of changes to an external battery performance such as the loss of a rated capacity, increased temperature rise during operation, or internal changes such as degradation due to corrosion.

[0037] In some embodiments, the term effective discharge resistance may include or be substituted with effective discharge impedance or effective discharge conductance, for example.

In the case of impedance, dividing voltage by current might include complex components (e.g., D7220 real and imaginary components). In the case of effective discharge conductance, the current is divided by the voltage.

[0038] In some embodiments, resistance or impedance may refer to the degree to which an electrochemical cell, module or pack prevents electrical current from moving through it for a particular voltage. Resistance or impedance might refer to ohmic resistance, for example, from conductors, bus bars, module or pack hardware, etc., or from diffusion resistance, double layer resistance, charge transfer resistance, which might reflect changes in mostly chemistry.

[0039] In step 330, since the effective discharge resistance changes with the health of the battery cells 130, the battery modules 110, and the battery pack 100 and changes with time as the battery cells 130, the battery modules 1 10, and the battery pack 100 ages, the determined effective discharge resistance can be used to determine the SoH states of one or more battery cells 130, one or more stacks of battery cells 130, one or more battery modules 110, and/or the battery pack 100.

[0040] In some embodiments, a table of effective discharge resistances and corresponding SoH states can be stored in the battery management system 120 (e.g., the processors 170 and/or the memory 180) for comparison, interpolation or extrapolation in view of the measured effective discharge resistance changes.

[0041] In some embodiments, the determined effective discharge resistance can compared with a reference effective discharge resistance (e.g., a beginning of life effective resistance) to calculate an SoH ratio, for example.

[0042] Referring to FIG. 6, an embodiment of a method for determining the SoH state is described with respect to a HEV, EV, PHEV, or REEV.

[0043] In step 340, wait for the battery cells 130, the battery modules 110, and/or the battery pack 100 in a HEV, EV, PHEV, or REEV to be in equilibrium condition and/or in a shutdown condition. In some embodiments, the process of calculating battery SoH from a D7220 resistance (e.g., an effective discharge resistance, an internal resistance, etc.) is performed when the vehicle is off-line so the battery is not under load and is as close to equilibrium as possible.

[0044] In step 350, the vehicle wakes itself up, for example, for a self-diagnostic, or is awakened by a user or a service technician. [0045] In step 360, the inverter 260 (e.g., power inverter) drains one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 via a constant current discharge over a period of time.

[0046] In step 370, during the constant drain discharge, the battery management system

120 can take one or more voltage measurements of the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 via the voltage sensors 200.

[0047] In step 380, an algorithm that is executed by the processor 170, the processor 230, and/or some other processor or circuit determines the effective resistance of the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 using the one or more voltage measurements. [0048] In step 390, the determine effective resistance can be compared to the effective resistance of the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 at the beginning of life (possibly adjusted for temperature) and an SoH ratio can be calculated.

[0049] In some cases, the end of life (EoL) of an electrochemical cell, module or pack can be defined via a defined and pre-established increase in resistance. Some embodiments can compensate for factors such as Irm_S, in particular, in applications in which the cycle is very restrictive or well-defined. Such a calculation can include a linear approximation based on a stored table, or can be based on a particular electrochemical cell, module or pack performance model, which can be based on a particular electrochemical theory or a particular physical cycle testing. D7220

[0050] In some embodiments, the above procedures can similarly be performed during charging of the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 for a period of time. Instead of discharging using the inverter 260, the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100 can be charged for a period of time by a charging device such as the plug-in charger 280 (e.g., a plug-in charger of a PHEV). The battery cell, module or pack is charged at a low current value to make the measurement. In an HEV, the existing power electronics of the vehicle can provide a variable load on the battery to make the resistance calculation— no extra external circuitry is required. The effective resistance and the SoH state can then be determined for the one or more of the battery cells 130, one or more of the battery modules 110, and/or the battery pack 100.

[0051] Some embodiments use existing battery management systems in existing HEVs,

EVs, PHEVs, or REEVs in which processors (e.g., hybrid supervisory control processors) of existing battery management systems can be programmed to execute the above-described operations.

[0052] Some embodiments provide that relatively simple calculations can be performed using existing components in vehicles (e.g., HEVs, EVs, PHEVs, or REEVs) to calculate the SoH state, which is typically a result of a complex and computationally-intensive algorithm. Some the existing components may include, for example, one or more of the following: electrochemical cells, modules or packs; inverters; voltage sensors; current sensors; timers; and plug-in chargers.

[0053] Some embodiments provide that the determined SoH can be used in applications for battery fuel gauging.

[0054] Some embodiments provide that the determined SoH can be used as an input to the control systems of the vehicle and can influence control system parameters such power limits, for example. D7220

[0055] Some embodiments facilitate an accurate prediction of energy storage system power limits in a vehicle through temperature and age.

[0056] Some embodiments provide for an algorithm used to estimate resistance without creation and development of an extensive cell, unit, module or pack electrochemical model, an extensive electrochemical analysis and testing, and without adding additional hardware detection mechanisms.

[0057] Some embodiments use effective discharge resistance directly in estimating battery SoH and neglects polarization resistance (PR) and cold cranking amperes (CCA) of a battery cell, unit, module or pack. [0058] Some embodiments provide that the effective discharge resistances are the only parameters needed to estimate battery SoH.

[0059] Some embodiments use internal resistance (IR) directly in estimating battery SoH and neglect PR and CCA of a battery cell, unit, module or pack.

[0060] Some embodiments use a vehicular hybrid system to measure parameters and to calculate battery SoH without the use of external circuits to excite the battery, and without extra microcomputer bandwidth and processing time.

[0061] Some embodiments use the determined SoH to optimize battery performance and/or to mitigate safety hazards, for example.

[0062] Some embodiments provide an algorithm methodology that addresses a problem in calculating a combined Ohmic resistance and charge transfer resistance for the life of an energy storage system of a vehicle in an relatively accurate and predictable manner without expanding to an equivalent circuit model or extending calibration mechanisms.

[0063] Some embodiments solve the problem of the undetermined SoH status of an electrochemical cell, module or pack and the unexpected performance due to the undetermined SoH status. Some of the systems and methods described herein are independent of supplier or D7220 battery management system capability to measure individual electrochemical parameters through hardware, software and/or firmware algorithms in a non-destructive manner.

[0064] Some embodiments provide that the determined battery SoH can be used to provide a State of Life (SoL) or SoH indicator for Rechargeable Energy Storage Systems (RESS) such as batters for PHEV, REEV, and other types of electric vehicles.

[0065] Some embodiments contemplate being in compliance with EPA regulation

86.1806-01, which is incorporated by reference herein in its entirety. A portion of the regulation notes that the manufacturer equip "off-vehicle charge capable HEVs" with a useful life indicator for the battery system consisting of a light that must illuminate the first time the battery system is unable to achieve an all-electric operating range (starting from a full state-of-charge) which is at least 75 percent of the range determined by for the vehicle in the Urban Driving Schedule portion of the All-Electric Range Test (see the California Zero-Emission and Hybrid Electric Vehicle Exhaust Emission Standards and Test Procedures for 2003 and Subsequent Model Year Passenger Cars, Light-Duty Trucks and Medium Duty Vehicles." Thus, some embodiments contemplate using the determined battery SoH state or status as a basis for a battery-system- useful-life indicator as set forth in the regulation.

[0066] Some embodiments also contemplate similarly using the determined battery SoH state or status as a basis for such a useful life indicator for medium duty and heavy duty electric and hybrid cars, vehicles or trucks that use, for example, electrochemical cells, modules or packs.

[0067] Some embodiments provide that the determined battery SoH state or status can be used to predict the remaining life of the battery cell, module or pack.

[0068] Such a prediction can be used to hold suppliers accountable for longer warranty commitments with respect to electrical vehicles (EVs). For example, EV batteries are currently warrantied for 3.5 years, which is less than the warranty for other vehicle components. The reason is that suppliers are reluctant to specify or commit to a timeframe and are conservative D7220 with their warranty commitments in which the supplier defines the end of a battery's lifetime (EoL) as a point at which capacity decreases by 20-25 percent from beginning of life (BoL). This is typically because there a multitude of factors that decrease capacity, most of which are outside the control of the battery suppliers (e.g., drive cycles, peak-to-peak current draw, charge/discharge frequency, plug-in charge operation, charge depletion v. charge sustaining modes, environmental factors, time in use v. time in storage, and the nature of the battery cell construction and electrochemistry).

[0069] Using the battery SoH determination described herein, a longer warranty period could be agreed to with suppliers on the conditions that battery SoH is monitored and maintained below a certain level during a durability test phase of a project. This would give suppliers the confidence to take additional warranty risk based on the durability test results.

Claims

D7220 CLAIMS

1. A battery management system for use with one or more batteries, comprising: a processor; and

sensors coupled to the processor and coupled to the one or more batteries,

wherein the processor causes the one or more batteries to be drained over a period of time, wherein the sensors take one or more voltage measurements of the one or more batteries during the period of time, wherein the processor determines an effective resistance of the one or more batteries based on the one or more voltage measurements, and wherein the processor determines a battery state-of-health of the one or more batteries based on the determined effective resistance.

2. The battery management system of claim 1, wherein the processor causes an inverter to drain the one or more batteries at a constant current.

3. The battery management system of claim 1, wherein the battery management system is part of a hybrid electric vehicle, an electric vehicle, a plug-in electric vehicle, or a range-extended electric vehicle.

4. The battery management system of claim 1, wherein the one or more batteries include one or more lithium-ion batteries.

5. The battery management system of claim 1, wherein the period of time is a programmable period of time or a preset period of time.

6. The battery management system of claim 1, wherein the processor determines a battery state-of-health of the one or more batteries by performing a comparison between the determined effective resistance and a reference effective resistance. D7220

7. The battery management system of claim 1, wherein the reference effective resistance corresponds to the effective resistance of the one or more batteries at the beginning of battery life.

8. The battery management system of claim 1, wherein the reference effective resistance comprises one or more resistances stored in a table residing in a memory that is coupled to the processor, and wherein the table shows a correspondence between the one or more resistance values and respective battery state-of-health values.

9. A method of managing one or more batteries, comprising:

causing one or more batteries to be drained over a period of time;

taking one or more voltage measurements of the one or more batteries during the period of time;

determining an effective resistance of the one or more batteries based on the one or more voltage measurements; and

determining a battery state-of-health of the one or more batteries based on the determined effective resistance.

10. The method of claim 9, wherein the step of causing one or more batteries to be drained includes causing one or more batteries to be drained over the period of time at a constant current.

11. The method of claim 9, wherein the step of causing one or more batteries to be drained includes causing one or more batteries to be drained in a hybrid electric vehicle, an electric vehicle, a plug-in electric vehicle, or a range-extended electric vehicle. D7220

12. The method of claim 9, wherein the one or more batteries include one or more lithium-ion batteries.

13. The method of claim 9, wherein the period of time is a programmable period of time or a preset period of time.

14. The method of claim 9, comprising:

replacing the one or more batteries based on the determined battery state-of-health.

15. A battery management system for use with one or more batteries, comprising: means for causing one or more batteries to be drained over a period of time;

means for taking one or more voltage measurements of the one or more batteries during the period of time;

means for determining an effective resistance of the one or more batteries based on the one or more voltage measurements; and

means for determining a battery state-of-health of the one or more batteries based on the determined effective resistance.