Systems and methods for contactless power converter aging detection

Patent metadata

Field	Value
Publication number	US-12313447-B2
Application number	US-202217813827-A
Country	US
Kind code	B2
Filing date	Jul 20, 2022
Priority date	Jul 26, 2021
Publication date	May 27, 2025
Grant date	May 27, 2025

Abstract

Systems, methods, and computer-readable media are disclosed for contactless power converter aging detection. An example method may include receiving, from a microphone, first acoustic data associated with a power converter. The example method may also include converting the first acoustic data into second acoustic data, wherein the first acoustic data is time domain data and the second acoustic data is frequency domain data. The example method may also include determining, by a machine learning model and based on the second acoustic data, a remaining useful life value associated with the power converter.

First claim

What is claimed is:

1. A method comprising:
receiving first acoustic data associated with a power converter, wherein the first acoustic data is captured by a microphone;
converting the first acoustic data into second acoustic data, wherein the first acoustic data is time domain data and the second acoustic data is frequency domain data;
converting the second acoustic data into a first voltage/current (V/I) spectrum;
receiving an ideal V/I spectrum for the power converter;
comparing the first V/I spectrum to the ideal V/I spectrum;
determining, by a machine learning model and based on the comparison, a remaining useful life value associated with the power converter; and
automatically sending, based on determining the remaining useful life value associated with the power converter, a signal to the power converter to shut down or modify an operational parameter of the power converter.

2. The method of claim 1 , wherein comparing the first V/I spectrum to the ideal V/I spectrum further comprises determining a variation between a first gate threshold voltage associated with the first V/I spectrum and a second gate threshold voltage associated with the ideal V/I spectrum.

3. The method of claim 1 , wherein the first acoustic data is produced by switching events of power semiconductors associated with the power converter.

4. The method of claim 1 , further comprising:
receiving a second V/I spectrum associated with the power converter;
comparing the first V/I spectrum and the second V/I spectrum; and
training the machine learning model based on a difference between the first V/I spectrum and the second V/I spectrum.

5. The method of claim 4 , wherein the second V/I spectrum is received from a spectrum analyzer.

6. The method of claim 1 , wherein the microphone is associated with a device that is separate from the power converter or a standalone microphone that is separate from the power converter, wherein the microphone is placed at a distance from the power converter.

7. A system comprising:
a microphone;
a power converter;
at least one processor; and
at least one memory storing computer-executable instructions, that when executed by the at least one processor, cause the at least one processor to:
receive first acoustic data associated with the power converter, wherein the first acoustic data is captured by the microphone;
convert the first acoustic data into second acoustic data, wherein the first acoustic data is time domain data and the second acoustic data is frequency domain data;
convert the second acoustic data into a first voltage/current (V/I) spectrum;
receive an ideal V/I spectrum for the power converter;
compare the first V/I spectrum to the ideal V/I spectrum;
determine, by a machine learning model and based on the comparison, a remaining useful life value associated with the power converter; and
automatically send, based on determining the remaining useful life value associated with the power converter, a signal to the power converter to shut down or modify an operational parameter of the power converter.

8. The system of claim 7 , wherein comparing the first V/I spectrum to the ideal V/I spectrum further comprises determining a variation between a first gate threshold voltage associated with the first V/I spectrum and a second gate threshold voltage associated with the ideal V/I spectrum.

9. The system of claim 7 , wherein the first acoustic data is produced by switching events of power semiconductors associated with the power converter.

10. The system of claim 7 , wherein the computer-executable instructions further cause the at least one processor to:
receiving a second V/I spectrum associated with the power converter;
comparing the first V/I spectrum and the second V/I spectrum; and
training the machine learning model based on a difference between the first V/I spectrum and the second V/I spectrum.

11. The system of claim 10 , wherein the second V/I spectrum is received from a spectrum analyzer.

12. The system of claim 7 , wherein the microphone is associated with a device that is separate from the power converter or a standalone microphone that is separate from the power converter, wherein the microphone is placed at a distance from the power converter.

13. A non-transitory computer-readable medium storing computer-executable instructions, that when executed by at least one processor, cause the at least one processor to:
receive first acoustic data associated with a power converter, wherein the first acoustic data is captured by a microphone;
convert the first acoustic data into second acoustic data, wherein the first acoustic data is time domain data and the second acoustic data is frequency domain data;
convert the second acoustic data into a first voltage/current (V/I) spectrum;
receive an ideal V/I spectrum for the power converter;
compare the first V/I spectrum to the ideal V/I spectrum;
determine, by a machine learning model and based on the comparison, a remaining useful life value associated with the power converter; and
automatically send, based on determining the remaining useful life value associated with the power converter, a signal to the power converter to shut down or modify an operational parameter of the power converter.

14. The non-transitory computer-readable medium of claim 13 , wherein comparing the first V/I spectrum to the ideal V/I spectrum further comprises determining a variation between a first gate threshold voltage associated with the first V/I spectrum and a second gate threshold voltage associated with the ideal V/I spectrum.

15. The non-transitory computer-readable medium of claim 13 , wherein the first acoustic data is produced by switching events of power semiconductors associated with the power converter.

16. The non-transitory computer-readable medium of claim 13 ,
wherein the computer-executable instructions further cause the at least one processor to:
receiving a second V/I spectrum associated with the power converter;
comparing the first V/I spectrum and the second V/I spectrum; and
training the machine learning model based a difference between the first V/I spectrum and the second V/I spectrum.

17. The non-transitory computer-readable medium of claim 16 , wherein the second V/I spectrum is received from a spectrum analyzer.

Assignees

• Univ Florida State Res Found Inc

Inventors

• Shi Yanjun
• Liu Xiuwen

Classifications

• H04R1/08

  Mouthpieces; {Microphones;} Attachments therefor · CPC title

• H04R1/028

  associated with devices performing functions other than acoustics, e.g. electric candles · CPC title

• G01H3/08 Primary

  Analysing frequencies present in complex vibrations, e.g. comparing harmonics present {(acoustic presence detection G01V1/001)} · CPC title