Method and system for detecting symptoms and determining an optimal remedy pattern for a faulty device

• US 10,223,188 B2
• Filed: 05/19/2015
• Issued: 03/05/2019
• Est. Priority Date: 05/09/2012
• Status: Active Grant

• Current Assignee: Infosys Limited
• Original Assignee: Infosys Limited
• Inventors: Agnihotram, Gopichand, Pisipati, Radha Krishna
• Primary Examiner(s): Chu, Gabriel
• Application Number:
  US14/715,775
• Publication Number:
  US 20150309855A1
• Time in Patent Office: 1,386 Days
• CPC Class Codes: G05B 23/02 Electric testing or monitoring
• CPC Class Codes: G05B 23/0262 Confirmation of fault detection, e.g. extra checks to confirm that a failure has indeed occurred
• CPC Class Codes: G05B 23/0286 Modifications to the monitored process, e.g. stopping operation or adapting control
• CPC Class Codes: G06F 11/0739 in a data processing system embedded in automotive or aircraft systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G07C 5/0808 Diagnosing performance data testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005

Abstract

A method, non-transitory computer readable medium, and anomaly detection computing apparatus that detects one or more of a plurality of symptoms associated with a device when the device is operating in a fault mode. One or more of a plurality of components of the device that are potentially faulty are identified based on the detected symptoms. One or more tests are performed on each of the one or more of the components to confirm that at least a subset of the one or more of the components is faulty. An optimal remedy pattern is determined for the subset of the one or more of the components based at least in part on historical data and cost data obtained from a database, and the optimal remedy pattern is output.

Claims

1. A method for optimizing symptom detection and performing an optimal remedy pattern for faulty device components, the method comprising:

obtaining, by a computing device, control sample data for a device when the device is operating in a no-fault mode and sensor data for each of a plurality of the components of the device when the device is operating in a fault mode;

performing, by the computing device, an exploratory data analysis on the control sample data and the sensor data to generate optimized control sample data and sensor data;

generating, by the computing device, a multivariate probability distribution from the optimized sensor data, wherein one or more parameters of the multivariate probability distribution are estimated by a maximum-likelihood estimation technique;

detecting, by the computing device, one or more of a plurality of symptoms associated with one or more of the plurality of the components of the device based on a comparison of the optimized sensor data to the optimized control sample data using a likelihood ratio test and the multivariate probability distribution;

identifying, by the computing device, one or more of the plurality of components of the device that are potentially faulty based on the detected one or more of the plurality of symptoms;

performing, by the computing device, one or more tests on each of the one or more of the plurality of components that are potentially faulty to confirm that at least a subset of the one or more of the components that are potentially faulty is faulty and to determine a status of the device;

determining, by the computing device, an optimal remedy pattern for the subset of the one or more of the components based at least in part on historical data and cost data obtained from a database; and

performing, by the computing device, the optimal remedy pattern to remedy one of the subset of the one or more of the components.

2. The method of claim 1, further comprising:

dynamically updating, by the computing device, the device based on the remediation of the one or more components according to the optimal remedy pattern;

determining, by the computing device, when the device is still operating in the fault mode after the remedying the one of the subset of the one or more of the components; and

repeating, by the computing device, the remedying for another of the subset of the one or more of the components based on a new remedial plan when the determining indicates that the device is still operating in the fault mode.

3. The method of claim 1, wherein the historical data comprises a correspondence of the components with the plurality of symptoms, one or more remedy patterns corresponding to the plurality of symptoms, or a correspondence of the remedy patterns with the components, the method further comprising:

updating, by the computing device, the historical data based on the optimal remedy pattern after the remediation.

4. The method of claim 1, wherein the cost data comprises a renew cost, an observation cost for the components, an observation cost of the device, or an investigation cost.

5. The method of claim 1, wherein the symptoms are detected based on the sensor data from the one or more sensor devices associated with the plurality of the components of the device using one or more measurement error models.

6. The method of claim 1, further comprising generating, by the computing device, at least one of remedy cost data without testing or remedy cost data with testing based on the cost data, wherein the optimal remedy pattern is further determined based on the remedy cost data without testing or the remedy cost data with testing.

7. A non-transitory computer readable medium having stored thereon instructions for optimizing symptom detection and performing an optimal remedy pattern for faulty device components comprising executable code which when executed by a processor, causes the processor to perform steps comprising:

obtaining control sample data for a device when the device is operating in a no-fault mode and sensor data for each of a plurality of the components of the device when the device is operating in a fault mode;

performing an exploratory data analysis on the control sample data and the sensor data to generate optimized control sample data and sensor data;

generating a multivariate probability distribution from the optimized sensor data, wherein one or more parameters of the multivariate probability distribution are estimated by a maximum-likelihood estimation technique;

detecting one or more of a plurality of symptoms associated with one or more of the plurality of the components of the device based on a comparison of the optimized sensor data to the optimized control sample data using a likelihood ratio test and the multivariate probability distribution;

identifying one or more of the plurality of components of the device that are potentially faulty based on the detected one or more of the plurality of symptoms;

performing one or more tests on each of the one or more of the plurality of components that are potentially faulty to confirm that at least a subset of the one or more of the components that are potentially faulty is faulty and to determine a status of the device;

determining an optimal remedy pattern for the subset of the one or more of the components based at least in part on historical data and cost data obtained from a database; and

performing the optimal remedy pattern to remedy one of the subset of the one or more of the components.

8. The non-transitory computer readable medium of claim 7, wherein the executable code when executed by the processor further causes the processor to perform at least one additional step comprising:

dynamically updating the device based on the remediation of the one or more components according to the optimal remedy pattern;

determining when the device is still operating in the fault mode after the remedying the one of the subset of the one or more of the components; and

repeating the remedying for another of the subset of the one or more of the components based on a new remedial plan when the determining indicates that the device is still operating in the fault mode.

9. The non-transitory computer readable medium of claim 7, wherein the historical data comprises a correspondence of the components with the plurality of symptoms, one or more remedy patterns corresponding to the plurality of symptoms, or a correspondence of the remedy patterns with the components, wherein the executable code when executed by the processor further causes the processor to perform at least one additional step comprising:

updating the historical data based on the optimal remedy pattern after the remediation.

10. The non-transitory computer readable medium of claim 7, wherein the cost data comprises a renew cost, an observation cost for the components, an observation cost of the device, or an investigation cost.

11. The non-transitory computer readable medium of claim 7, wherein the symptoms are detected based on the sensor data from the one or more devices associated with the plurality of the components of the device using one or more measurement error models.

12. The non-transitory computer readable medium of claim 7, wherein the executable code when executed by the processor further causes the processor to perform at least one additional step comprising generating at least one of remedy cost data without testing or remedy cost data with testing based on the cost data, wherein the optimal remedy pattern is further determined based on the remedy cost data without testing or the remedy cost data with testing.

13. A symptom detection and analysis computing device, comprising at least one processor and a memory coupled to the processor which is configured to be capable of executing programmed instructions comprising and stored in the memory to:

obtain control sample data for a device when the device is operating in a no-fault mode and sensor data for each of a plurality of the components of the device when the device is operating in a fault mode;

perform an exploratory data analysis on the control sample data and the sensor data to generate optimized control sample data and sensor data;

generate a multivariate probability distribution from the optimized sensor data, wherein one or more parameters of the multivariate probability distribution are estimated by a maximum-likelihood estimation technique;

detect one or more of a plurality of symptoms associated with one or more of the plurality of the components of the device based on a comparison of the optimized sensor data to the optimized control sample data using a likelihood ratio test and the multivariate probability distribution;

identify one or more of a plurality of components of the device that are potentially faulty based on the detected one or more of the plurality of symptoms;

perform one or more tests on each of the one or more of the plurality of components that are potentially faulty to confirm that at least a subset of the one or more of the components that are potentially faulty is faulty and to determine a status of the device;

determine an optimal remedy pattern for the subset of the one or more of the components based at least in part on historical data and cost data obtained from a database; and

perform the optimal remedy pattern to remedy one of the subset of the one or more of the components.

14. The symptom detection and analysis computing device of claim 13, wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction to:

dynamically update the device based on the remediation of the one or more components according to the optimal remedy pattern;

determine when the device is still operating in the fault mode after the remedy of the one of the subset of the one or more of the components; and

repeat the remedy for another of the subset of the one or more of the components based on a new remedial plan when the determining indicates that the device is still operating in the fault mode.

15. The symptom detection and analysis computing device of claim 13, wherein the historical data comprises a correspondence of the components with the plurality of symptoms, one or more remedy patterns corresponding to the plurality of symptoms, or a correspondence of the remedy patterns with the components, wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction to:

update the historical data based on the optimal remedy pattern after the remediation.

16. The symptom detection and analysis computing device of claim 13, wherein the cost data comprises a renew cost, an observation cost for the components, an observation cost of the device, or an investigation cost.

17. The symptom detection and analysis computing device of claim 13, wherein the symptoms are detected based on the sensor data from the one or more devices associated with the plurality of the components of the device using one or more measurement error models.

18. The symptom detection and analysis computing device of claim 13, wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction to generate at least one of remedy cost data without testing or remedy cost data with testing based on the cost data, wherein the optimal remedy pattern is further determined based on the remedy cost data without testing or the remedy cost data with testing.