System and method for estimating long term characteristics of battery

• US 9,255,973 B2
• Filed: 08/21/2008
• Issued: 02/09/2016
• Est. Priority Date: 08/23/2007
• Status: Active Grant

• Current Assignee: LG Energy Solution Limited (LG Corporation)
• Original Assignee: LG Chem Limited (LG Corporation)
• Inventors: Song, Hyun-Kon, Cho, Jeong-Ju, Choo, Yeon-Uk, Son, Mi-Young, Lee, Ho-Chun
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): COUGHLAN, PETER D
• Application Number:
  US12/674,647
• Publication Number:
  US 20110191278A1
• Time in Patent Office: 2,728 Days
• Field of Search: 706/17
• US Class Current: 1/1
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables
• CPC Class Codes: G06N 7/00 Computing arrangements based on specific mathematical models

Abstract

A system for estimating long term characteristics of a battery includes a learning data input unit for receiving initial characteristic learning data and long term characteristic learning data of a battery to be a learning object; a measurement data input unit for receiving initial characteristic measurement data of a battery to be an object for estimation of long term characteristics; and an artificial neural network operation unit for receiving the initial characteristic learning data and the long term characteristic learning data from the learning data input unit to allow learning of an artificial neural network, receiving the initial characteristic measurement data from the measurement data input unit and applying the learned artificial neural network thereto, and thus calculating long term characteristic estimation data from the initial characteristic measurement data of the battery and outputting the long term characteristic estimation data.

Claims

1. A computer having a computer-readable medium on which a system for estimating long term characteristics of a battery is recorded, the system comprising:

a learning data input unit configured to receive a first plurality of initial charging capacity variation data sets for the battery, each initial charging capacity variation data set respectively obtained at a plurality of initial cycles when the battery has been periodically charged and discharged in a predetermined number of cycles, and receive a first long term charging capacity variation data set for the battery obtained at a predetermined latter long term cycle after the initial cycles during the predetermined number of cycles, wherein one cycle refers to one charging and one discharging of the battery and the charging capacity variation data includes a plurality of charging capacity values in accordance with varying of a charging voltage or a charging time;

a measurement data input unit configured to receive a second plurality of initial charging capacity variation data sets respectively measured during the initial cycles of the predetermined number of cycles for the battery to be an object for estimating a second long term charging capacity variation data set predicted at the predetermined latter long term cycle of the predetermined number of cycles;

an artificial neural network operation unit configured to receive the first plurality of initial charging capacity variation data sets and the first long term charging capacity data set from the learning data input unit to allow learning of an artificial neural network, receive the measured second plurality of initial charging capacity variation data sets from the measurement data input unit and apply the learned artificial neural network thereto, and thus determine the second long term charging capacity variation data set predicted at the predetermined latter long term cycle of the predetermined number of cycles from the measured second plurality of the initial charging capacity variation data sets; and

a display device configured to output the determined second long term charging capacity variation data set.

2. The computer according to claim 1, wherein the learned artificial neural network has at least one neuron layer arranged in series, and

wherein the neuron layer converts an input vector into an output vector corresponding to the second long term charging capacity variation data set, such that a bias vector and a weight matrix calculated by the learning of the artificial neural network are reflected on the input vector, the input vector on which the bias vector and the weight matrix are reflected is processed by a neuron transfer function, and then a result of the neuron transfer function is output as the output vector.

3. The computer according to claim 2, wherein, in the serial arrangement of the neuron layer, a first neuron layer has an input vector composed of the second plurality of initial charging capacity variation data sets.

4. The computer according to claim 1, further comprising an initial characteristic measurement sensor for measuring the second plurality of initial charging capacity variation data sets for the battery put into an activating process and then outputting the measured second plurality of initial charging capacity variation data sets,

wherein the measurement data input unit receives the measured second plurality of initial charging capacity variation data sets from the initial characteristic measurement sensor.

5. The computer according to claim 1, further comprising a long term characteristic evaluation unit for receiving the second long term charging capacity variation data set from the artificial neural network operation unit and determining a long term characteristic quality by comparing the received second long term charging capacity variation data set with a criterion long term charging capacity variation data set.

6. The computer according to claim 5, wherein the long term characteristic evaluation unit outputs a long term characteristic, quality determination result of the battery in a graphic-user interface through the display device.

7. A method for estimating long term characteristics of a battery, the method comprising:

(a) receiving, by a learning data input unit, a first plurality of initial charging capacity variation data sets for the battery, each initial charging capacity variation data set respectively obtained at a plurality of initial cycles when the battery has been periodically charged and discharged in a predetermined number of cycles, and receiving, by the learning data input unit, a first long term charging capacity variation data set for the battery obtained at a predetermined latter long term cycle after the initial cycles during the predetermined number of cycles, wherein one cycle refers to one charging and one discharging cycle of the battery and the charging capacity variation data includes a plurality of charging capacity values in accordance with varying of a charging voltage or a charging time;

(b) receiving, by a measurement data input unit, a second plurality of initial charging capacity variation data sets respectively measured during the initial cycles of the predetermined numbers of cycles for the battery to be an object for estimating a second long term charging capacity variation data set predicted at the predetermined latter long term cycle of the predetermined number of cycles;

(c) receiving, by an artificial neural network operation unit, the first plurality of initial charging capacity variation data sets and the first long term charging capacity variation data set from the learning data input unit to allow learning of an artificial neural network, receiving, by the artificial neural network operation unit, the measured second plurality of initial charging capacity variation data sets from the measurement data input unit and applying the learned artificial neural network thereto, and thus determining, by the artificial neural network operation unit, the second long term charging capacity variation data set predicted at the predetermined latter cycle of the predetermined number of cycles from the measured second plurality of initial charging capacity variation data sets; and

(d) outputting, by a display device, the determined second long term charging capacity variation data set.

8. The method according to claim 7, wherein the learned artificial neural network has at least one neuron layer arranged in series, and

wherein the step (c) includes:

(c1) converting the measured second plurality of initial charging capacity variation data sets into an input vector;

(c2) inputting the input vector into a first neuron layer of the neuron layer arrangement;

(c3) each neuron layer of the neuron layer arrangement reflecting a bias vector and a weight matrix calculated by the learning of the artificial neural network on the input vector and then processing the input vector by a neuron transfer function such that the input vector is converted into an output vector corresponding to the second long term charging capacity variation data set; and

(c4) a last neuron layer of the neuron layer arrangement outputting the output vector.

9. The method according to claim 7, wherein the step (c) includes:

putting the battery to be an object for long term characteristic estimation into the battery activating process; and

obtaining the second plurality of initial charging capacity variation data set for the battery by using an initial characteristic measurement sensor.

10. The method according to claim 7, further comprising:

comparing the second long term charging capacity variation data set with a criterion long term charging capacity variation data set to determine a long term characteristic quality of the battery.

11. The method according to claim 10, further comprising:

visually displaying the determined long term characteristic quality of the battery.