Fault diagnosis using distributed PCA architecture

• US 10,650,616 B2
• Filed: 04/06/2018
• Issued: 05/12/2020
• Est. Priority Date: 04/06/2018
• Status: Active Grant

• Current Assignee: University of Connecticut (State of Connecticut), Toyota Jidosha Kabushiki Kaisha (Toyota Motor Corporation)
• Original Assignee: University of Connecticut (State of Connecticut)
• Inventors: Zhang, Lingyi, Chen, Weiqiang, Pattipatti, Krishna, Bazzi, Ali M., Joshi, Shailesh N., Dede, Ercan M.
• Primary Examiner(s): Marc, McDieunel
• Application Number:
  US15/947,662
• Publication Number:
  US 20190311552A1
• Time in Patent Office: 767 Days
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G07C 5/008 communicating information to a remotely located station transmission systems for measured values G08C
• CPC Class Codes: G07C 5/0808 Diagnosing performance data testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005
• CPC Class Codes: G07C 5/0816 Indicating performance data, e.g. occurrence of a malfunction

Abstract

A system includes a vehicle having an electronic device, a sensor designed to detect sensor data corresponding to at least one property of the electronic device, an output device designed to output data, and a vehicle network access device designed to transmit the sensor data. The system also includes a machine learning server separate from the vehicle and having a machine learning processor designed to receive the sensor data, and generate, using a machine learning algorithm, a model of the electronic device. The machine learning processor is also designed to determine that a fault is likely to occur with the electronic device by conducting a T squared statistical analysis of the sensor data using the model, and generate a signal to be transmitted to the vehicle network access device when the fault is likely to occur and output information indicating that the fault is likely to occur.

Claims

1. A system for analyzing a health status of a component of vehicles, comprising:

a vehicle having;

an electronic device,

a sensor configured to detect sensor data corresponding to at least one property of the electronic device,

an output device configured to output data, and

a vehicle network access device configured to transmit the sensor data; and

a machine learning server separate from the vehicle and having a machine learning processor configured to;

receive the sensor data,

generate, using a machine learning algorithm, a model of the electronic device,

determine that a fault is likely to occur with the electronic device by conducting a T squared statistical analysis of the sensor data using the model, and

generate a signal to be transmitted to the vehicle network access device when the fault is likely to occur such that the output device of the vehicle outputs information indicating that the fault is likely to occur.

2. The system of claim 1 wherein the machine learning processor is further configured to determine that the model is faulty by conducting a Q analysis of the sensor data, and to generate a new model or update the model of the electronic device when the model is faulty.

3. The system of claim 1 wherein the machine learning processor is configured to receive the sensor data corresponding to multiple electronic devices of multiple vehicles and to generate the model of the electronic device using the sensor data corresponding to the multiple electronic devices of the multiple vehicles.

4. The system of claim 3 wherein the machine learning processor is further configured to determine that the model of the electronic device is faulty by conducting a Q analysis of the sensor data corresponding to the multiple electronic devices of the multiple vehicles, and to generate a new model or update the model of the electronic device when the model is faulty.

5. The system of claim 3 wherein the machine learning processor is further configured to at least one of:

generate a separate model of the electronic device for each of the multiple vehicles;

or

generate a single model of the electronic device for use with each of the multiple vehicles.

6. The system of claim 3 wherein the machine learning processor is further configured to generate a first model of the electronic device for a first group of the multiple vehicles that each include a first type of vehicle, and to generate a second model of the electronic device for a second group of the multiple vehicles that each include a second type of vehicle.

7. The system of claim 6 wherein the first type of vehicle includes at least one of a vehicle make, a vehicle model, a geographical area in which the first group of the multiple vehicles are used, a weather condition experienced by the first group of the multiple vehicles, or a sensor drift of the first group of the multiple vehicles.

8. The system of claim 1 wherein the machine learning algorithm is a principal component analysis (PCA) algorithm.

9. The system of claim 1 wherein the sensor data includes at least one of a cycling current, a temperature, a thermal resistance, a power step, a voltage level, or a current level.

10. A system for analyzing a health status of a component of a vehicle, comprising:

a plurality of vehicles each having;

an electronic device,

a sensor configured to detect sensor data corresponding to at least one property of the electronic device,

an output device configured to output data, and

a vehicle network access device configured to transmit the sensor data of the electronic device; and

a machine learning server separate from each of the plurality of vehicles and having a machine learning processor configured to;

receive the sensor data of the electronic device for each of the plurality of vehicles,

generate, using a machine learning algorithm, a model of the electronic device,

determine that a fault is likely to occur with the electronic device of at least one of the plurality of vehicles by conducting a T squared statistical analysis of the sensor data using the model, and

generate a signal to be transmitted to the vehicle network access device of the at least one of the plurality of vehicles when the fault is likely to occur such that the output device of the at least one of the plurality of vehicles outputs information indicating that the fault is likely to occur.

11. The system of claim 10 wherein the machine learning processor is further configured to determine that the model is faulty by conducting a Q analysis of the sensor data, and to generate a new model or update the model of the electronic device when the model is faulty.

12. The system of claim 10 wherein the machine learning processor is further configured to determine that the model of the electronic device is faulty by conducting a Q analysis of the sensor data corresponding to the electronic device of each of the plurality of vehicles, and to generate a new model or update the model of the electronic device when the model is faulty.

13. The system of claim 10 wherein the machine learning processor is further configured to at least one of:

generate a separate model of the electronic device for each of the plurality of vehicles;

or

generate a single model of the electronic device for use with each of the plurality of vehicles.

14. The system of claim 10 wherein the machine learning processor is further configured to generate a first model of the electronic device for a first group of the plurality of vehicles that each include a first type of vehicle, and to generate a second model of the electronic device for a second group of the plurality of vehicles that each include a second type of vehicle.

15. The system of claim 14 wherein the first type of vehicle includes at least one of a vehicle make, a vehicle model, a geographical area in which the first group of the plurality of vehicles are used, a weather condition experienced by the first group of the plurality of vehicles, or a sensor drift of the first group of the plurality of vehicles.

16. A method for analyzing a health status of a component of vehicles, comprising:

detecting, by a sensor coupled to or positioned near an electronic device of a vehicle, sensor data corresponding to at least one property of the electronic device;

transmitting, by a vehicle network access device of the vehicle, the sensor data of the electronic device;

receiving, by a machine learning processor of a machine learning server being separate from the vehicle, the sensor data of the electronic device;

generating, by the machine learning processor, a model of the electronic device;

determining, by the machine learning processor, that a fault is likely to occur with the electronic device by conducting a T squared statistical analysis of the sensor data using the model; and

outputting, by an output device of the vehicle, information indicating that the fault is likely to occur.

17. The method of claim 16 further comprising:

determining, by the machine learning processor, that the model is faulty by conducting a Q analysis of the sensor data; and

generating or updating, by the machine learning processor, the model of the electronic device when the model is faulty.

18. The method of claim 16 further comprising:

receiving, by the machine learning processor, the sensor data corresponding to multiple vehicles; and

generating, by the machine learning processor, the model of the electronic device using the sensor data corresponding to the electronic device of each of the multiple vehicles.

19. The method of claim 18 further comprising at least one of:

generating, by the machine learning processor, a separate model of the electronic device for each of the multiple vehicles;

or

generating, by the machine learning processor, a single model of the electronic device for use with each of the multiple vehicles.

20. The method of claim 18 further comprising:

generating, by the machine learning processor, a first model of the electronic device for a first group of the multiple vehicles that each include a first type of vehicle; and

generating, by the machine learning processor, a second model of the electronic device for a second group of the multiple vehicles that each include a second type of vehicle.