Model-based estimation of battery hysteresis

• US 7,957,921 B2
• Filed: 02/19/2008
• Issued: 06/07/2011
• Est. Priority Date: 02/19/2008
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Zhang, Xiaodong, Tang, Xidong, Frisch, Damon R., Koch, Brian J., Chin, Yuen-Kwok
• Primary Examiner(s): TSAI, CAROL S W
• Application Number:
  US12/033,197
• Publication Number:
  US 20090210179A1
• Time in Patent Office: 1,204 Days
• Field of Search: 702/63, 702/64, 702/65, 702/183, 320/132, 320/136
• US Class Current: 702/63
• CPC Class Codes: H01M 10/48 Accumulators combined with arrangements for measuring, testing or indicating the condition of cells, e.g. the level or density of the electrolyte constructional details of current conducting connections for detecting conditions inside cells or batteries, e.g. details of voltage sensing terminals, H01M50/569
• CPC Class Codes: Y02E 60/10 Energy storage using batteries

Abstract

A method of battery state of charge estimation considering battery hysteresis includes using a Preisach-model-based algorithm to calculate a battery state of charge.

Claims

1. A method of battery state of charge estimation for a battery within a vehicle considering battery hysteresis behavior in an on-board vehicle system comprising:

executing within the on-board vehicle system an algorithm based on the Preisach-model to calculate the battery state of charge as a function of received data inputs comprising an open circuit voltage and a temperature measured by a temperature sensor, executing the algorithm comprising;

defining a plurality of elementary hysterons considering minimum and maximum values of the state of charge for said battery, wherein the elementary hysterons describe the battery hysteresis behavior and each of the elementary hysterons comprises a binary output depending upon historical operation of the battery based upon the Preisach model;

training a plurality of sets of elementary hysteron specific weights at a plurality of temperatures;

utilizing the plurality of elementary hysterons;

selecting one of the sets of elementary hysteron specific weights based upon the measured temperature; and

utilizing the selected set of elementary hysteron specific weights, wherein one of the elementary hysteron specific weights of the selected set is utilized for each of the plurality of elementary hysterons modifying an effect of each of the binary outputs; and

applying the received data inputs to said algorithm;

wherein the executing the algorithm within the on-board vehicle system comprises utilizing the plurality of elementary hysterons and the elementary hysteron specific weights stored in memory of the on-board vehicle system.

2. The method of claim 1, wherein the elementary hysteron specific weights are initially trained.

3. A method of battery state of charge estimation for a battery within a vehicle considering battery hysteresis behavior in an on-board vehicle system comprising:

executing within the on-board vehicle system an algorithm based on the Preisach-model to calculate the battery state of charge as a function of received data inputs comprising an open circuit voltage and a temperature measured by a temperature sensor, executing the algorithm comprising;

defining a plurality of elementary hysterons considering minimum and maximum values of the state of charge for said battery, wherein the elementary hysterons describe the battery hysteresis behavior and each of the elementary hysterons comprises a binary output depending upon historical operation of the battery based upon the Preisach model;

training a plurality of sets of elementary hysteron specific weights at a plurality of temperatures;

utilizing the plurality of elementary hysterons;

selecting one of the sets of elementary hysteron specific weights based upon the measured temperature;

utilizing the selected set of elementary hysteron specific weights, wherein one of the elementary hysteron specific weights of the selected set is utilized for each of the plurality of elementary hysterons modifying an effect of each of the binary outputs; and

applying the received data inputs to said algorithm;

wherein the executing the algorithm within the on-board vehicle system comprises utilizing the plurality of elementary hysterons and the elementary hysteron specific weights stored in memory of the on-board vehicle system;

wherein the elementary hysteron specific weights are initially trained; and

wherein said training comprises processing test data gathered by cycling a test battery, said cycling operative to capture a boundary hysteresis loop and internal hysteresis loops.

4. The method of claim 3, wherein said open circuit voltage is estimated from parameters available while said battery is in a state of charge or discharge.

5. The method of claim 4, wherein the parameters comprise a battery voltage and a battery current.

6. A method of battery state of charge estimation for a battery within a vehicle considering battery hysteresis behavior in an on-board vehicle system comprising:

executing within the on-board vehicle system an algorithm based on the Preisach-model to calculate the battery state of charge as a function of received data inputs comprising an open circuit voltage, executing the algorithm comprising;

defining a basic hysteresis curve characterized in said Preisach-model comprising a plurality of elementary hysterons describing the battery hysteresis behavior, wherein each of the elementary hysterons describes a binary output depending upon historical operation of the battery based upon the Preisach model;

weighting said basic hysteresis curve with elementary hysteron specific weights based upon training data, wherein one of the elementary hysteron specific weights is utilized for each of the plurality of elementary hysterons modifying an effect of each of the binary outputs;

shifting said basic hysteresis curve in order to match an output range of said algorithm to a range of actual battery states of charge; and

applying the received data inputs to said algorithm, comprising;

utilizing the plurality of elementary hysterons; and

utilizing the elementary hysteron specific weights;

wherein the executing the algorithm within the on-board vehicle system comprises utilizing the plurality of elementary hysterons and the elementary hysteron specific weights stored in memory of the on-board vehicle system.

7. The method of claim 6, wherein said training data comprises test data gathered by cycling a test battery, said cycling operative to capture a boundary hysteresis loop and internal hysteresis loops.

8. The method of claim 6, wherein said range of actual battery states of charge comprise a maximum value corresponding to a state of charge for a fully charged battery and a minimum value corresponding to a state of charge for a fully discharged battery.

9. The method of claim 8, wherein said open circuit voltage is estimated from parameters available while said battery is in a state of charge or discharge.

10. The method of claim 9, wherein the parameters include a battery voltage, a battery current, and a temperature measured by a temperature sensor.