Efficient battery tester

• US 10,295,611 B2
• Filed: 04/20/2016
• Issued: 05/21/2019
• Est. Priority Date: 06/09/2015
• Status: Active Grant

• Current Assignee: Premier Technologies Pty. Ltd.
• Original Assignee: Premier Technologies Pty. Ltd.
• Inventors: Cheng, Chun Sing, Lau, Wing Hong, Chung, Shu Hung Henry
• Primary Examiner(s): Bachner, Robert G
• Application Number:
  US15/134,025
• Publication Number:
  US 20160363634A1
• Time in Patent Office: 1,126 Days
• CPC Class Codes: G01R 31/3647 for determining the ability of a battery to perform a critical function, e.g. cranking
• CPC Class Codes: G01R 31/3648 comprising digital calculation means, e.g. for performing an algorithm
• CPC Class Codes: G01R 31/385 Arrangements for measuring battery or accumulator variables for monitoring G01R31/382
• CPC Class Codes: G01R 31/389 Measuring internal impedance, internal conductance or related variables
• CPC Class Codes: G01R 31/392 Determining battery ageing or deterioration, e.g. state of health

Abstract

A battery tester for efficiently extracting intrinsic characteristics of a battery. The battery tester comprises a sensing unit, a controller, bi-directional power converter, and an energy storage unit so that energy discharged from a battery during testing is stored in the energy storage unit and recycled back to the battery. The battery tester uses a high current pulse test and a switching excitation test with a modified particle arm optimization algorithm to analyze precise model parameters that can be used to describe the health conditions of a battery.

Claims

1. A battery tester for evaluating intrinsic characteristics of a battery, comprising:

a. at least one sensor for sensing voltage and current;

b. an energy storage unit for temporarily storing power from the battery, wherein the energy storage unit comprises a supercapacitor;

c. a bi-directional power converter for facilitating energy transfer between the battery and the energy storage unit;

d. a controller for controlling the bi-directional power converter, the controller comprising a processor, a non-transitory computer readable storage medium, and instructions stored in the non-transitory computer readable storage medium, which when executed by the processor cause the processor to perform a battery test method, wherein the battery test method comprises transferring energy from the battery to the energy storage unit and transferring energy from the energy storage unit to the battery using the bi-directional power converter; and

e. a display, wherein a user interface on the display provides controls operable for selecting the battery test method;

f. wherein the battery test method is selected from a first test mode for quick evaluation of the battery or a second test mode for precise evaluation of the battery, wherein the first test mode comprises a high current pulse test and the second test mode comprises the high current pulse test and a switching excitation test;

g. wherein the high current pulse test comprises drawing a high current from the battery to the energy storage unit over a short period of time, returning energy from the energy storage unit to the battery, and measuring voltage and current changes using the at least one sensor;

h. wherein the switching excitation test comprises transferring energy from the battery to the energy storage unit and transferring energy from the energy storage unit to the battery at a fixed frequency, and measuring voltage and current using the at least one sensor;

i. wherein the first test mode determines battery internal resistance, cold cranking ampere (“

CCA”

), and state of charge (“

SOC”

), wherein battery internal resistance is determined by evaluating

2. The battery tester of claim 1, wherein the battery tester is installed in an automobile to perform an analysis of an automobile battery.

3. The battery tester of claim 1, wherein the battery tester is a standalone device operable to connect with a standalone battery and display test results of the standalone battery.

4. The battery tester of claim 1, wherein the battery tester further comprises a power supply, whereby the power supply supplies energy to the battery tester in order to test a deeply discharged battery.

5. A battery tester for evaluating intrinsic characteristics of a battery, comprising:

a. at least one sensor for sensing voltage and current;

b. an energy storage unit for temporarily storing power from the battery, wherein the energy storage unit comprises a supercapacitor;

c. a bi-directional power converter for facilitating energy transfer between the battery and the energy storage unit;

d. a controller for controlling the bi-directional power converter, the controller comprising;

i. a processor, a non-transitory computer readable storage medium, and instructions stored in the non-transitory computer readable storage medium, which when executed by the processor cause the processor to perform at least one battery test method, wherein the at least one battery test method comprises transferring energy from the battery to the energy storage unit and transferring energy from the energy storage unit to the battery using the bi-directional power converter;

ii. wherein the battery tester further comprising a user interface on a display, wherein the user interface comprises controls for selecting the at least one battery test method;

iii. wherein the at least one battery test method comprises conducting a high current pulse test, wherein the high current pulse test comprises drawing a high current from the battery to the energy storage unit over a short period of time, returning energy from the energy storage unit to the battery, and measuring voltage and current changes using the at least one sensor;

iv. wherein the battery characteristics determined are battery internal resistance, cold cranking ampere (“

CCA”

), and state of charge (“

SOC”

), wherein battery internal resistance is determined by evaluating

6. The battery tester of claim 5, further comprising determining the intrinsic characteristics of the battery by using a recursive algorithm to increase accuracy of an electrical model of the battery, wherein the recursive algorithm comprises:

a. initializing a set of particles representative of voltage and current of the battery, wherein each particle has a velocity and a position;

b. updating the set of particles for a number of iterations, wherein updating the set of particles comprises;

i. determining an inaccuracy value of each particle using an inaccuracy function, wherein the inaccuracy function measures a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time, wherein a particle with a lowest inaccuracy value is considered a best particle of the set of particles, and wherein a particle with a highest inaccuracy value is considered a worst particle of the set of particles;

ii. generating a replacement particle by generating and injecting a perturbation into the best particle;

iii. evaluating an inaccuracy value of the replacement particle using the inaccuracy function;

iv. if the replacement particle has a lower inaccuracy value than the best particle, replacing the best particle with the replacement particle in the set of particles, or else replacing the worst particle with the replacement particle in the set of particles; and

v. updating the velocity and the position of each particle in the set of particles; and

c. determining state of charge, cold cranking ampere, and internal resistance of the electrical model of the battery based on the set of particles.

7. The battery tester of claim 6, wherein determining an inaccuracy value of each particle is determined by evaluating function ε

_T(P)=1/T∫

₀^T|v_b,act(t)−

v_b,m[t, P, i_b,act(t)]|dt, wherein ε

_T is a function of a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time T, wherein v_b,act(t) is an actual battery voltage measured at time t, wherein i_b,act(t) is an actual battery current measured at time t, wherein v_b,m[t, P, i_act(t)] is battery voltage at time t calculated using the electrical model with a parameter set P under the actual battery current i_b,act(t).

8. The battery tester of claim 7, wherein state of charge (“

SOC”

) at time t is determined by evaluating SOC(t)=SOC(0)+1/Q₀∫

₀^ti_b(θ

)dθ

, wherein SOC(0) is the state of charge at a beginning of a period, Q₀ is fully charge capacity of the battery, time i_b(θ

) is the battery current at time θ

.

9. A battery tester for evaluating intrinsic characteristics of a battery, comprising:

a. at least one sensor for sensing voltage and current;

b. an energy storage unit for temporarily storing power from the battery, wherein the energy storage unit comprises a supercapacitor;

c. a bi-directional power converter for facilitating energy transfer between the battery and the energy storage unit;

d. a controller for controlling the bi-directional power converter, the controller comprising;

i. a processor, a non-transitory computer readable storage medium, and instructions stored in the non-transitory computer readable storage medium, which when executed by the processor cause the processor to perform at least one battery test method, wherein the at least one battery test method comprises transferring energy from the battery to the energy storage unit and transferring energy from the energy storage unit to the battery using the bi-directional power converter;

ii. wherein the battery tester further comprising a user interface on a display, wherein the user interface comprises controls for selecting the at least one battery test method;

iii. wherein the at least one battery test method comprises conducting a high current pulse test, wherein the high current pulse test comprises drawing a high current from the battery to the energy storage unit over a short period of time, returning energy from the energy storage unit to the battery, and measuring voltage and current changes using the at least one sensor; and

iv. wherein the at least one battery test method further comprises conducting a switching excitation test, wherein the switching excitation test comprises transferring energy from the battery to the energy storage unit and transferring energy from the energy storage unit to the battery at a fixed frequency, and measuring voltage and current using the at least one sensor.

10. The battery tester of claim 9, further comprising determining the intrinsic characteristics of the battery by using a recursive algorithm to increase accuracy of an electrical model of the battery, wherein the recursive algorithm comprises:

a. initializing a set of particles representative of voltage and current of the battery, wherein each particle has a velocity and a position;

b. updating the set of particles for a number of iterations, wherein updating the set of particles comprises;

i. determining an inaccuracy value of each particle using an inaccuracy function, wherein the inaccuracy function measures a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time, wherein a particle with a lowest inaccuracy value is considered a best particle of the set of particles, and wherein a particle with a highest inaccuracy value is considered a worst particle of the act of particles;

ii. generating a replacement particle by generating and injecting a perturbation into the best particle,

iii. evaluating an inaccuracy value of the replacement particle using the inaccuracy function;

iv. if the replacement particle has a lower inaccuracy value than the best particle, replacing the best particle with the replacement particle in the set of particles, or else replacing the worst particle with the replacement particle in the set of particles; and

v. updating the velocity and the position of each particle in the set of particles; and

c. determining state of charge, cold cranking ampere, and internal resistance of the electrical model of the battery based on the set of particles.

11. The battery tester of claim 10, wherein determining an inaccuracy value of each particle is determined by evaluating function

12. The battery tester of claim 11, wherein state of charge (“

SOC”

) at time t is determined by evaluating

13. A method of testing batteries for intrinsic characteristics, the method comprising:

at a circuit having a controller, a bi-directional power converter, an energy storage unit, and at least one sensor, wherein the energy storage unit comprises a supercapacitor;

a. performing a battery test method, wherein the battery test method comprises;

i. recording a voltage and current of a battery;

ii. directing the controller to transfer energy from the battery to the energy storage unit using the bi-directional power converter;

iii. directing the controller to transfer energy from the energy storage unit to the battery using the bi-directional power converter; and

iv. measuring voltage and current of the circuit using the at least one sensor;

b. determining intrinsic characteristics of the battery based on measured voltage and current of the circuit;

c. the method for testing batteries further comprising providing a user interface on a display, wherein the user interface provides a selection of battery test methods to perform on the battery;

d. wherein the selection of battery test methods comprises a first battery test method to simulate a high current pulse, wherein the step of directing the controller to transfer energy from the battery to the energy storage unit using the bi-directional power converter is conducted over a short period of time; and

e. wherein the intrinsic characteristics comprise battery internal resistance, cold cranking ampere (“

CCA”

), and state of charge (“

SOC”

), wherein battery internal resistance is determined by evaluating

14. The method for testing batteries of claim 13, wherein determining intrinsic characteristics of the battery comprises using a recursive algorithm to increase accuracy of an electrical model of the battery, wherein the recursive algorithm comprises:

a. Initializing a set of particles representative of voltage and current of the battery, wherein each particle has a velocity and a position;

b. updating the set of particles for a number of iterations, wherein updating the set of particles comprises;

i. determining an inaccuracy value of each particle using an inaccuracy function, wherein the inaccuracy function measures a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time, wherein a particle with a lowest inaccuracy value is considered a best particle of the set of particles, and wherein a particle with a highest inaccuracy value is considered a worst particle of the set of particles;

ii. generating a replacement particle by generating and injecting a perturbation into the best particle;

iii. evaluating an inaccuracy value of the replacement particle using the inaccuracy function;

iv. if the replacement particle has a lower inaccuracy value than the best particle, replacing the best particle with the replacement particle in the set of particles, or else replacing the worst particle with the replacement particle in the set of particles; and

v. updating the velocity and the position of each particle in the set of particles; and

c. determining state of charge, cold cranking ampere, and internal resistance of the electrical model of the battery based on the set of particles.

15. The method for testing batteries of claim 14, wherein determining an inaccuracy value of each particle is determined by evaluating function ε

_T(P)=1/T∫

₀^T|v_b,act(t)−

v_b,m[t, P, i_b,act(t)]|dt, wherein ε

_T is a function of a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time T, wherein v_b,act(t) is an actual battery voltage measured at time t, wherein i_b,act(t) is an actual battery current measured at time t, wherein v_b,m[t, P, i_act(t)] is battery voltage at time t calculated using the electrical model with a parameter set P under the actual battery current i_b,act(t).

16. The method for testing batteries of claim 15, wherein state of charge (“

SOC”

) at time t is determined by evaluating SOC(t)=SOC(0)+1/Q₀∫

₀^ti_b(θ

)dθ

, wherein SOC(0) is the state of charge at a beginning of a period, Q₀ is a fully charge capacity of the battery, and time i_b(θ

) is the battery current at time θ

.

17. A method of testing batteries for intrinsic characteristics, the method comprising:

at a circuit having a controller, a bi-directional power converter, an energy storage unit, and at least one sensor, wherein the energy storage unit comprises a supercapacitor;

a. performing a battery test method, wherein the battery test method comprises;

i. recording a voltage and current of a battery;

ii. directing the controller to transfer energy from the battery to the energy storage unit using the bi-directional power converter;

iii. directing the controller to transfer energy from the energy storage unit to the battery using the bi-directional power converter; and

iv. measuring voltage and current of the circuit using the at least one sensor;

b. determining intrinsic characteristics of the battery based on measured voltage and current of the circuit;

c. the method for testing batteries further comprising providing a user interface on a display, wherein the user interface provides a selection of battery test methods to perform on the battery;

d. wherein the selection of battery test methods comprises a first battery test method to simulate a high current pulse, wherein the step of directing the controller to transfer energy from the battery to the energy storage unit using the bi-directional power converter is conducted over a short period of time; and

e. wherein the selection of battery test methods comprises a second battery test method to simulate switching excitation, wherein the steps of directing the controller to transfer energy from the battery to the energy storage unit and from the energy storage unit to the battery are performed at a fixed frequency.

18. The method for testing batteries of claim 17, wherein determining intrinsic characteristics of the battery comprises using a recursive algorithm to increase accuracy of an electrical model of the battery, wherein the recursive algorithm comprises:

a. initializing a set of particles representative of voltage and current of the battery, wherein each particle has a velocity and a position;

b. updating the set of particles for a number of iterations, wherein updating the set of particles comprises;

i. determining an inaccuracy value of each particle using an inaccuracy function, wherein the inaccuracy function measures a difference between actual battery voltage-current characteristics and an electrical model with estimated intrinsic parameters for a period of time, wherein a particle with a lowest inaccuracy value is considered a best particle of the set of particles, and wherein a particle with a highest inaccuracy value is considered a worst particle of the set of particles;

ii. generating a replacement particle by generating and injecting a perturbation into the best particle;

iii. evaluating an inaccuracy value of the replacement particle using the inaccuracy function;

iv. if the replacement particle has a lower inaccuracy value than the best particle, replacing the best particle with the replacement particle in the set of particles, or else replacing the worst particle with the replacement particle in the set of particles; and

v. updating the velocity and the position of each particle in the set of particles; and

c. determining state of charge, cold cranking ampere, and internal resistance of the electrical model of the battery based on the set of particles.

19. The method for testing batteries of claim 18, wherein determining an inaccuracy value of each particle is determined by evaluating function

20. The method for testing batteries of claim 19, wherein state of charge (“

SOC”

) at time t is determined by evaluating