Battery model identification method

• US 8,655,612 B2
• Filed: 11/12/2009
• Issued: 02/18/2014
• Est. Priority Date: 12/01/2008
• Status: Active Grant

• Current Assignee: Calsonic Kansei Corporation
• Original Assignee: Calsonic Kansei Corporation
• Inventors: Itabashi, Kinnosuke
• Primary Examiner(s): HENSON, MISCHITA L
• Application Number:
  US13/131,348
• Publication Number:
  US 20110231124A1
• Time in Patent Office: 1,559 Days
• Field of Search: 702 63- 65, 320/106, 340/636.1, 34063612-63613, 340/636.15
• US Class Current: 702/63
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables
• CPC Class Codes: G01R 31/389 Measuring internal impedance, internal conductance or related variables
• CPC Class Codes: H01M 10/052 Li-accumulators
• CPC Class Codes: H01M 10/482 for several batteries or cells simultaneously or sequentially
• CPC Class Codes: Y02E 60/10 Energy storage using batteries

Abstract

In a battery model identification method of the present invention, an M-sequence input electric current production part inputs M-sequence signals with different frequency components as an electric current input into the battery. In this occasion, terminal voltage of the battery is measured by a voltage sensor, and a parameter estimation part executes system identification based on the measured result to calculate frequency characteristics of the battery. Resistance components and capacitance components as parameters of a battery model are identified based on the calculated frequency characteristics.

Claims

1. A battery model identification method of a battery capable of being discharged and charged, the method comprising the steps of:

setting parameters of a battery model of the battery, the parameters of the battery model including a plurality of resistance components and capacitance components;

inputting an electric current into the battery;

measuring a terminal voltage of the battery when the electric current is inputted;

system-identifying the battery model based on the electric current inputted to the battery and the measured terminal voltage;

calculating frequency characteristics of the battery; and

estimating the parameters of the battery model based on the calculated frequency characteristics of the battery,

wherein the electric current is produced by M-sequence signals with predetermined different frequency components,

wherein sampling frequencies utilized in said measuring of the terminal voltage are shifted according to a shift of the M-sequence signals, and

wherein said system-identifying of the battery model identifies the battery model (i) by extracting and synthesizing data on respective frequency responses in a predetermined frequency range and (ii) based on the data on the respective frequency responses.

2. The battery model identification method according to claim 1, further comprising the steps of:

calculating a state of charge (SOC) by accumulating an output current of the battery;

calculating open circuit voltage based on the SOC;

calculating overvoltage by subtracting the open circuit voltage from the terminal voltage of the battery; and

system-identifying the battery model using the overvoltage.

3. The battery model identification method according to claim 1, wherein the M-sequence signals are set such that (i) a set clock period of the M-sequence signals and a set sampling frequency have the same values and (ii) the predetermined frequency range of the M-sequence signals is set to be 1/100 to 1/10 times as wide as the set sampling frequency.

4. The battery model identification method according to claim 2, wherein the M-sequence signals are set such that (i) a set clock period of the M-sequence signals and a set sampling frequency have the same values and (ii) the predetermined frequency range of the M-sequence signals is set to be 1/100 to 1/10 times as wide as the set sampling frequency.