Method, the device, and the computer-readable recording medium for diagnosing the reason of the malfunction

Patent metadata

Field	Value
Publication number	US-12499656-B2
Application number	US-202318482030-A
Country	US
Kind code	B2
Filing date	Oct 6, 2023
Priority date	Jul 27, 2023
Publication date	Dec 16, 2025
Grant date	Dec 16, 2025

Abstract

A method for diagnosing a reason of a malfunction is provided. The method includes: receiving a signal to be diagnosed; decomposing the signal to be diagnosed into a plurality of sub-signals; transforming each of the plurality of sub-signals into a corresponding grayscale image; and inputting the corresponding grayscale images to a neural network model, and outputting a malfunction reason classification result through the neural network model. Accordingly, the method can be used for diagnosing the reason of the malfunction and solves the problem of incapable of diagnosing the reason of the malfunction. In addition, a device and a computer-readable recording medium for diagnosing the reason of the malfunction are also provided.

First claim

What is claimed is:

1 . A method for diagnosing a reason of a malfunction, comprising:
receiving a signal to be diagnosed; decomposing the signal to be diagnosed into a plurality of sub-signals; transforming each of the plurality of sub-signals into a corresponding grayscale image; and inputting the corresponding grayscale images to a neural network model, and outputting a malfunction reason classification result through the neural network model.

2 . The method according to claim 1 , wherein decomposing the signal to be diagnosed into a plurality of sub-signals is completed by an empirical mode decomposition model.

3 . The method according to claim 1 , wherein the sub-signals are intrinsic mode functions.

4 . The method according to claim 1 , wherein transforming each of the plurality of sub-signals into the corresponding grayscale image comprises:
performing signal transformation of the sub-signals by a Fourier transformation model; respectively mapping signal strength of the sub-signals after the signal transformation to a corresponding grayscale value; and integrating the corresponding grayscale values into the corresponding grayscale image.

5 . The method according to claim 1 , further comprising:
comparing the number of the sub-signals with a predetermined number of grayscale figures that the neural network model can receive; when the number of the sub-signals is greater than or equal to the predetermined number of grayscale figures, the decomposition of the signal to be diagnosed is stopped; and when the number of the sub-signals is less than the predetermined number of grayscale figures, at least one infilling image is provided to the neural network model.

6 . The method according to claim 5 , wherein the number of the at least one infilling image depends on the number of the sub-signals and the predetermined number of grayscale figures.

7 . The method according to claim 1 , wherein the neural network model is a convolutional neural network model trained by plural pieces of data, and each piece of the data comprises a machine system signal and a machine system state corresponding to the machine system signal.

8 . The method according to claim 1 , wherein the malfunction reason classification result comprises at least one of an unbalanced state, a non-parallel state, a shaft that is skew, and a loose state.

9 . A device for diagnosing a reason of a malfunction, suitable for connecting with an apparatus to be diagnosed by signal, in order to receive a signal to be diagnosed from the apparatus to be diagnosed, the device comprising:
a signal receiving unit, suitable for receiving the signal to be diagnosed; a processing unit, configured to couple with the signal receiving unit; and a storage unit, configured to couple with the processing unit, wherein the storage unit stores a code, and after the processing unit executes the code stored in the storage unit, the device can execute steps as described below: receiving a signal to be diagnosed; decomposing the signal to be diagnosed into a plurality of sub-signals; transforming each of the plurality of sub-signals into a corresponding grayscale image; and inputting the corresponding grayscale images to a neural network model, and outputting a malfunction reason classification result through the neural network model.

10 . The device according to claim 9 , wherein decomposing the signal to be diagnosed into a plurality of sub-signals is completed by an empirical mode decomposition model.

11 . The device according to claim 9 , wherein the sub-signals are intrinsic mode functions.

12 . The device according to claim 9 , wherein transforming each of the plurality of sub-signals into the corresponding grayscale image comprises:
performing signal transformation of the sub-signals by a Fourier transformation model; respectively mapping signal strength of the sub-signals after the signal transformation to a corresponding grayscale value; and integrating the corresponding grayscale values into the corresponding grayscale image.

13 . The device according to claim 9 , wherein after the processing unit executes the code stored in the storage unit, the device can execute steps as described below further comprising:
comparing the number of the sub-signals with a predetermined number of grayscale figures that the neural network model can receive; when the number of the sub-signals is greater than or equal to the predetermined number of grayscale figures, the decomposition of the signal to be diagnosed is stopped; and when the number of the sub-signals is less than the predetermined number of grayscale figures, at least one infilling image is provided to the neural network model.

14 . The device according to claim 13 , wherein the number of the at least one infilling image depends on the number of the sub-signals and the predetermined number of grayscale figures.

15 . A non-transitory computer-readable recording medium capable of diagnosing a reason of a malfunction, after a computer loads and executes a code stored in the non-transitory computer-readable recording medium, the non-transitory computer-readable recording medium can complete steps as described below:
receiving a signal to be diagnosed; decomposing the signal to be diagnosed into a plurality of sub-signals; transforming each of the plurality of sub-signals into a corresponding grayscale image; and inputting the corresponding grayscale images to a neural network model, and outputting a malfunction reason classification result through the neural network model.

16 . The non-transitory computer-readable recording medium according to claim 15 , wherein decomposing the signal to be diagnosed into a plurality of sub-signals is completed by an empirical mode decomposition model.

17 . The non-transitory computer-readable recording medium according to claim 15 , wherein the sub-signals are intrinsic mode functions.

18 . The non-transitory computer-readable recording medium according to claim 15 , wherein transforming each of the plurality of sub-signals into the corresponding grayscale image comprises:
performing signal transformation of the sub-signals by a Fourier transformation model; respectively mapping signal strength of the sub-signals after the signal transformation to a corresponding grayscale value; and integrating the corresponding grayscale values into the corresponding grayscale image.

19 . The non-transitory computer-readable recording medium according to claim 15 , wherein after the computer executes the code stored in the non-transitory computer-readable recording medium, the non-transitory computer-readable recording medium can complete steps as described below further comprising:
comparing the number of the sub-signals with a predetermined number of grayscale figures that the neural network model can receive; when the number of the sub-signals is greater than or equal to the predetermined number of grayscale figures, the decomposition of the signal to be diagnosed is stopped; and when the number of the sub-signals is less than the predetermined number of grayscale figures, at least one infilling image is provided to the neural network model.

20 . The non-transitory computer-readable recording medium according to claim 19 , wherein the number of the at least one infilling image depends on the number of the sub-signals and the predetermined number of grayscale figures.

Assignees

• Adlink Tech Inc

Inventors

• Tu Wei-Jyun
• Chen Yu-Yen
• Lin Chien-Chung

Classifications

• G06T5/10

  using non-spatial domain filtering · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T7/0004

  Industrial image inspection · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06F2218/12

  Classification; Matching · CPC title