Pattern recognition approach to battery diagnosis and prognosis

• US 8,972,213 B2
• Filed: 06/27/2008
• Issued: 03/03/2015
• Est. Priority Date: 06/27/2008
• 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, Zhang, Yilu, Chin, Yuen-Kwok
• Primary Examiner(s): Feliciano, Eliseo Ramos
• Assistant Examiner(s): SUAREZ, FELIX E
• Application Number:
  US12/147,556
• Publication Number:
  US 20090326841A1
• Time in Patent Office: 2,440 Days
• Field of Search: 702/30, 702/57, 702/58, 702 63- 65, 702/140, 702/179, 702/183, 702/185, 318/139, 320/132, 324/427, 324/433
• 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 31/392 Determining battery ageing or deterioration, e.g. state of health

Abstract

A method is provided for determining a state-of-health of a battery in a vehicle-during an engine cranking phase. An engine cranking phase is initiated. Characteristic data is recorded that includes battery voltage data and engine cranking speed data during the engine cranking phase. The characteristic data is provided to a pre-processing unit. The pre-processing unit normalizes the characteristic data for processing within a classifier. The normalized data is input o the classifier for determining the vehicle battery state-of-health. The classifier has a trained state-of-health decision boundary resulting from a plurality of trials in which predetermined characterization data is collected with known classes. The battery state-of-health is classified based on the trained state-of-health decision boundary.

Claims

1. A method for determining a state-of-health of a battery during an engine cranking phase, the method comprising the steps of:

initiating the engine cranking phase;

recording characteristic data including battery voltage data and engine cranking speed data during the engine cranking phase;

providing characteristic data to a pre-processing unit, the pre-processing unit normalizing the characteristic data for processing within a classifier;

inputting the normalized data to the classifier for determining the battery state-of-health, the classifier having a trained state-of-health decision boundary resulting from a plurality of trials in which predetermined characterization data is collected with known classes; and

classifying the battery state-of-health based on the trained state-of-health decision boundary.

2. A method of claim 1, wherein the trained state-of-health decision boundary is generated by the process comprising:

monitoring a plurality of batteries having known classes during respective engine cranking phases;

recording a battery voltage and engine cranking speed data for each monitored battery during the respective engine cranking phases;

normalizing the recorded battery voltage and engine cranking speed data of each monitored battery;

inputting the normalized data for each monitored battery into a training program, the training program learning a decision boundary in response to the normalized data of each monitored battery provided to the training program for establishing the classifier to distinguish each respective class; and

generating the trained state-of-health decision boundary in response to the normalized data of each monitored battery classified in the training program, the state-of-health decision boundary being configured for use in the classifier for classifying the vehicle battery state-of-health to a respective class, wherein at least one of the respective classes is associated with the state-of-health of a vehicle battery being below an acceptable level.

3. The method of claim 2 wherein inputting the normalized data to the training program includes inputting the normalized data to a neural network-based training program.

4. The method of claim 2 wherein inputting the normalized data to the training program includes inputting the normalized data to a support vector-based training program.

5. The method of claim 2 wherein recording the battery voltage and engine cranking speed data of each monitored battery during each engine cranking phase is obtained during an interval between a time the battery voltage decreases due to a starter engagement and the time when an engine speed reaches a predetermined engine speed.

6. The method of claim 2 wherein temperature data and state of charge data obtained during each engine cranking phase for each monitored battery are provided to the training program.

7. The method of claim 1 wherein temperature data is provided to the classifier for determining the respective class of a respective battery.

8. The method of claim 1 wherein a state of charge data is provided to the classifier for determining the respective class of a respective battery.

9. The method of claim 1 wherein the classifier outputs a percentage based battery state-of-health.

10. The method of claim 1 wherein the characteristic data is obtained during an interval between a time the battery voltage decreases due to a starter engagement and the time when an engine speed reaches a predetermined engine speed.

11. A battery state-of-health monitoring system comprising:

a preprocessing unit adapted to be coupled to sensed signals for obtaining characteristic data including battery voltage data and engine cranking speed data during an engine cranking phase for a starting of a vehicle; and

a classifier having a trained state-of-health decision boundary resulting from a plurality of trials in which predetermined characterization data is collected with known classes, the classifier further having a processor for comparing the characterization data and the trained state-of-health decision boundary to produce a state-of-health class indicating the degree of similarity with the known classes.

12. The system of claim 11 wherein the trained state-of-health decision boundaries resulting from a plurality of trials is generated by an training program, the plurality of trials include monitoring a plurality of batteries and associated engine speeds during respective engine cranking phases for collecting the predetermined characteristic data, wherein the collected predetermined characteristic data is used to establish pattern recognition within each class.

13. The system of claim 12 wherein the predetermined characteristic data further includes environmental data, the environmental data being provided to the training program for generating the state-of-health decisions boundary.

14. The system of claim 13 wherein the predetermined characteristic data further includes temperature data provided to the training program.

15. The system of claim 13 wherein the predetermined characteristic data further includes state of charge data provided to the training program.

16. The system of claim 11 wherein environmental factors including temperature data are provided to the classifier for determining the class of the battery state-of-health.

17. The system of claim 11 wherein environmental factors including state of charge data are provided to the classifier for determining the class of the battery state-of-health.

18. The system of claim 11 wherein the indicator is a display identifying a percentage based state-of-health of the battery.

19. The system of claim 11 wherein the indicator is a warning identifying the state-of-health of a battery is below an acceptable level.