FAILURE PREDICTION APPARATUS FOR ELECTRIC DEVICE AND FAILURE PREDICTION METHOD USING THE SAME

• US 20170206459A1
• Filed: 01/09/2017
• Published: 07/20/2017
• Est. Priority Date: 01/20/2016
• Status: Active Grant

• Current Assignee: Ulsan National Institute of Science and Technology
• Original Assignee: Ulsan National Institute of Science and Technology
• Inventors: KWON, Dae Il
• Granted Patent: US 10,713,580 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/00 Error detection; Error correction; Monitoring error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

Provided is a failure prediction method using a failure prediction apparatus, including: receiving time-series data about measured performance parameters from a device under test; encoding the time-series data with a plurality of symbols corresponding to a predetermined range; calculating a transition probabilities between the symbols of the encoded time-series data, and generating a transition matrix according to the transition probabilities; calculating an abnormal indicator, which is a difference between the transition matrix and a pre-stored database, and an increased value of the abnormal indicator; and comparing the increased value of the abnormal indicator and a predetermined threshold value, and if the increased value of the abnormal indicator is greater than the predetermined threshold value, predicting that failure of the device under test is to occur.

Claims

1. A failure prediction method using a failure prediction apparatus, comprising:

receiving time-series data about measured performance parameters from a device under test;

encoding the time-series data with a plurality of symbols corresponding to a predetermined range;

calculating a transition probabilities between the symbols of the encoded time-series data, and generating a transition matrix according to the transition probabilities;

calculating an abnormal indicator, which is a difference between the transition matrix and a pre-stored database, and an increased value of the abnormal indicator; and

comparing the increased value of the abnormal indicator and a predetermined threshold value, and if the increased value of the abnormal indicator is greater than the predetermined threshold value, predicting that failure of the device under test is to occur.

2. The method of claim 1, wherein

the database is generated by

encoding the time-series data about the performance parameters measured when the device under test is normally operated, and forming a transition matrix of the transition probabilities between the respective symbols.

3. The method of claim 1, wherein

the calculating of the abnormal indicator and the increased value of the abnormal indicator includes

calculating, using a matrix norm, the abnormal indicator, which is the difference between the transition matrix and the pre-stored database.

4. The method of claim 1, wherein

the increased value of the abnormal indicator includes

at least one of an instantaneous slope of the abnormal indicator and the number of continuous times the abnormal indicator is increased.

5. The method of claim 4, wherein

the calculating of the abnormal indicator and the increased value of the abnormal indicator includes

calculating the instantaneous slope of the abnormal indicator using the least square method.

6. A failure prediction apparatus, comprising:

an input unit configured to receive time-series data about measured performance parameters from a device under test;

an encoding unit configured to encode the time-series data with a plurality of symbols corresponding to a predetermined range;

a generating unit configured to calculate transition probabilities between the symbols of the encoded time-series data and generate a transition matrix according to the transition probabilities;

a calculating unit configured to calculate an abnormal indicator, which is a difference between the transition matrix and a pre-stored database, and an increased value of the abnormal indicator; and

a predicting unit configured to compare the increased value of the abnormal indicator and a predetermined threshold value, and if the increased value of the abnormal indicator is greater than the predetermined threshold value, predict that failure of the device under test is to occur.

7. The apparatus of claim 6, wherein

the database is generated by

encoding the time-series data about the performance parameters measured when the device under test is normally operated, and forming a transition matrix of the transition probabilities between the respective symbols.

8. The apparatus of claim 6, wherein

the calculating unit is configured to

calculate the abnormal indicator, which is the difference between the transition matrix and the pre-stored database, using a matrix norm.

9. The apparatus of claim 6, wherein

the increased value of the abnormal indicator includes

at least one of an instantaneous slope of the abnormal indicator and the number of continuous times the abnormal indicator is increased.

10. The apparatus of claim 9, wherein

the calculating unit is configured to

calculate the instantaneous slope of the abnormal indicator using the least square method.