Vibration anomaly detection system and method

Abstract

A vibration anomaly detection method is suitable for an electronic component, and comprises the following steps: a processing device obtains a plurality of frames of detection images obtained by shooting the electronic component, obtains a first vibration parameter of the electronic component according to a plurality of blocks corresponding to the electronic component in the plurality of frames of detection images, and obtains a second vibration parameter of the electronic component according to the first vibration parameter; the processor is used for acquiring a first vibration parameter generated by the vibration sensing element when the vibration sensing element senses the vibration of the electronic element, acquiring a second vibration parameter generated by the vibration sensing element when the vibration sensing element senses the vibration of the electronic element, judging whether the first vibration parameter and the second vibration parameter conform to respective preset ranges or not, and outputting a warning notice when judging that any one of the first vibration parameter and the second vibration parameter does not conform to respective preset ranges.

Description

Technical Field

[0001] The present invention relates to a vibration anomaly detection system and method. Background Art

[0002] In electronic devices, vibrations of electronic components are ubiquitous. However, irregular vibrations may cause damage to electronic components and affect the normal operation of the devices. Currently, the detection methods for vibrations of electronic components include using software for simulation analysis.

[0003] However, using software for simulation analysis cannot perform real-time and online detection, and has certain limitations. Moreover, using software for simulation analysis may also obtain inaccurate results due to inaccurate parameter settings. Summary of the Invention

[0004] In view of the above, the present invention provides a vibration anomaly detection system and method to solve the above problems.

[0005] A vibration anomaly detection method according to an embodiment of the present invention is applicable to an electronic component, and the method includes the following steps executed by a processing device: obtaining a plurality of frames of detection images obtained by photographing the electronic component; obtaining a first vibration parameter of the electronic component according to a plurality of blocks corresponding to the electronic component in the plurality of frames of detection images; obtaining a second vibration parameter generated by a vibration sensing element sensing the vibration of the electronic component; determining whether the first vibration parameter and the second vibration parameter meet their respective preset ranges; and when it is determined that any one of the first vibration parameter and the second vibration parameter does not meet its respective preset range, outputting a warning notification.

[0006] A vibration anomaly detection system according to an embodiment of the present invention is applicable to an electronic component, and includes: a photographing element, a sensing element, and a processing device. The photographing element is used to photograph the electronic component to generate a plurality of frames of detection images. The sensing element is used to sense the vibration of the electronic component. The processing device is connected to the photographing element and the sensing element. The processing device is used to obtain a first vibration parameter according to a plurality of blocks corresponding to the electronic component in the plurality of frames of detection images, obtain a second vibration parameter from the sensing element, determine whether the first vibration parameter and the second vibration parameter meet their respective preset ranges, and output a warning notification when it is determined that any one of the first vibration parameter and the second vibration parameter does not meet its respective preset range.

[0007] Through the vibration anomaly detection system and method of the above one or more embodiments, problematic electronic components can be identified and warned in real time and accurately, and can be used for online detection. Moreover, through the auxiliary detection of the vibration sensor, more comprehensive and accurate vibration information can be provided, which helps to more accurately determine whether there is an anomaly in the electronic component.

[0008] The above description of the present disclosure and the following description of the embodiments are used to demonstrate and explain the concepts and principles of the present invention, and to provide a further explanation of the scope of the patent application of the present invention. Description of the Drawings

[0009] Figure 1 is a block diagram of a vibration anomaly detection system shown according to an embodiment of the present invention.

[0010] Figure 2 is a schematic diagram showing the installation of a photographic element and an electronic element shown according to an embodiment of the present invention.

[0011] Figure 3 is a flowchart of a vibration anomaly detection method shown according to an embodiment of the present invention.

[0012] Figure 4 and Figure 5 are schematic diagrams respectively showing normal vibration and abnormal vibration of an electronic element.

[0013] Figure 6 is a block diagram of a vibration anomaly detection system shown according to another embodiment of the present invention.

[0014] Figure 7 is a flowchart of training a detection model shown according to an embodiment of the present invention.

[0015] Figure 8 is a flowchart of retraining a detection model shown according to an embodiment of the present invention.

[0016] Description of the Reference Numerals:

[0017] 1, 2: Vibration anomaly detection system

[0018] 11, 21: Photographic element

[0019] 12, 22: Sensing element

[0020] 13, 23: Processing device

[0021] 24: Parameter database

[0022] A1: Circuit board A1

[0023] A11, A12: Electronic element

[0024] B1, B2, B3, B4: Block

[0025] S101, S103, S105, S107, S109, S201, S203, S301, S303: Step Detailed Description of the Embodiments

[0026] The following describes in detail the detailed features and advantages of the present invention in the embodiments. The content is sufficient for those skilled in the art to understand the technical content of the present invention and implement it accordingly. And according to the content, claims and drawings disclosed in this specification, those skilled in the art can easily understand the related purposes and advantages of the present invention. The following embodiments further illustrate the viewpoints of the present invention in detail, but do not limit the scope of the present invention in any way.

[0027] Please refer to Figure 1 and Figure 2 together, where Figure 1 is a block diagram of a vibration abnormality detection system shown according to an embodiment of the present invention. Figure 2 is a schematic diagram of the installation of a photographic element and an electronic element shown according to an embodiment of the present invention. The vibration abnormality detection system 1 is applicable to electronic elements. Taking Figure 2 as an example, the vibration abnormality detection system 1 can be used to detect the circuit board A1 and / or one or more of the electronic elements A11 and A12 on the circuit board A1. The vibration abnormality detection system 1 includes a photographic element 11, a sensing element 12, and a processing device 13. The electronic elements A11 and A12 can be electronic elements of different types and sizes, such as capacitors, inductors, resistors, and / or integrated circuits, etc., which are not limited in the present invention. The following takes the electronic element A11 as an example for illustration.

[0028] The photographic element 11 is used to photograph the electronic element A11 to generate multiple frames of detection images, where the photographic element 11 can be a high-speed camera. The sensing element 12 is used to sense the electronic element A11 to generate a vibration result, where the sensing element 12 can be a pressure sensor or a displacement sensor. The processing device 13 is electrically or communicatively connected to the photographic element 11 and the sensing element 12. The processing device 13 can control the shooting frequency of the photographic element 11 and the sampling rate of the sensing element 12. The processing device 13 is used to judge whether the vibration state of the electronic element A11 is abnormal according to the detection image and the sensing result of the sensing element 12. The processing device 13 may include one or more processors, and the processors are, for example, a central processing unit, a graphics processing unit, a microcontroller, a programmable logic controller, or other processors with signal processing functions.

[0029] To illustrate the operation of the vibration abnormality detection system 1 in more detail, please refer to Figure 1 and Figure 3 together, where Figure 3 is a flowchart of a vibration abnormality detection method shown according to an embodiment of the present invention. As Figure 3As shown, the vibration anomaly detection method includes: Step S101: Obtain multiple frames of detection images obtained by photographing an electronic component; Step S103: Obtain a first vibration parameter of the electronic component according to multiple blocks corresponding to the electronic component in the multiple frames of detection images; Step S105: Obtain a second vibration parameter generated by a vibration sensing component sensing the vibration of the electronic component; Step S107: Determine whether the first vibration parameter and the second vibration parameter meet their respective preset ranges; when the determination result of Step S107 is "yes", the vibration anomaly detection method can be ended; and when the determination result of Step S107 is "no", execute Step S109: Output a warning notification. It should be particularly noted that Figure 3 Step S103 is shown to be executed before Step S105, and Step S103 can also be executed after Step S105, or Step S103 and Step S105 are executed simultaneously.

[0030] In Step S101, the processing device 13 obtains multiple frames of detection images from the photographing component 11, and the detection images are obtained by the photographing component 11 photographing the electronic component.

[0031] In Step S103, the processing device 13 obtains a first vibration parameter of the electronic component according to multiple blocks corresponding to the electronic component in the detection images. The first vibration parameter can be at least one of vibration amplitude, vibration frequency, and displacement. To illustrate Step S103 in detail, please refer to Figure 4 and Figure 5 , where Figure 4 and Figure 5 are schematic diagrams showing the normal vibration and abnormal vibration of the electronic component respectively. In Figure 4 and Figure 5 , the solid line is the contour of the stationary electronic component in the detection image, and the dotted line is the contour of the vibrating electronic component in the detection image.

[0032] Taking Figure 4 as an example, one frame of detection image includes a block B1, and another frame of detection image includes another block B2. In Step S103, the processing device 13 takes the absolute value of the displacement of the blocks B1 and B2 relative to the solid line contour as the first vibration parameter, or takes the absolute value of the displacement of the blocks B3 and B4 relative to the solid line contour as the first vibration parameter. It should be particularly noted that the processing device 13 can take the sum of the absolute values of the displacements as the first vibration parameter, or take the average value of the absolute values of the displacements as the first vibration parameter.

[0033] In Step S105, the processing device 13 obtains a second vibration parameter generated by sensing the vibration of the electronic component from the sensing component 12. Depending on the type of the sensing component 12, the second vibration parameter can be at least one of vibration amplitude, vibration frequency, and displacement.

[0034] In step S107, the processing device 13 determines whether the first vibration parameter conforms to the corresponding preset range and whether the second vibration parameter conforms to the corresponding preset range. For Figure 4 and Figure 5 example, the first vibration parameter is displacement. When the result of the detected image is as Figure 4 shown, the first vibration parameter obtained according to blocks B1 and B2 conforms to the corresponding preset range, and the vibration anomaly detection method can be ended; when the result of the detected image is as Figure 5 shown, the first vibration parameter obtained according to blocks B3 and B4 does not conform to the corresponding preset range, and the processing device 13 executes step S109 to output a warning notification.

[0035] In another example, the first vibration parameter is vibration frequency, the second vibration parameter is displacement, the preset range corresponding to the first vibration parameter is, for example, 5 kilohertz (5 kHz), 0.5 kHz, 50 Hz, etc., and there can be an error range of 5%; the preset range corresponding to the second vibration parameter can be 0.1 millimeter, and there can be an error range of 5%. The preset range can be obtained according to the specification of the electronic component or the average value of multiple measurement values. Therefore, when the preset range is set to 5 kHz and the first vibration parameter is 8 kHz, and / or the second vibration parameter is 0.15 millimeter, it indicates that the electronic component may be abnormal, and the processing device 13 executes step S109 to output a warning notification.

[0036] In addition, in an embodiment, the processing device 13 can store a detection model. In step S107, the processing device 13 can input the first vibration parameter and the second vibration parameter into the detection model to obtain the detection result output by the detection model, where the detection result indicates whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges. The detection model can be a trained discriminative artificial intelligence model, such as a support vector machine (SVM), a decision tree, a logistic regression, etc. In other words, the processing device 13 can store the trained detection model, and when the detection result indicates that both the first vibration parameter and the second vibration parameter conform to their respective preset ranges, end the vibration anomaly detection method; and when the detection result indicates that at least one of the first vibration parameter and the second vibration parameter does not conform to the corresponding preset range, execute step S109.

[0037] Through the vibration anomaly detection system and method of one or more of the above embodiments, problematic electronic components can be identified and warned in real time and accurately, and can be used for online detection.

[0038] Please refer to Figure 6 where Figure 6It is a block diagram of a vibration anomaly detection system shown in another embodiment of the present invention. The vibration anomaly detection system 2 is applicable to electronic components and includes a photographing element 21, a sensing element 22, a processing device 23, and a parameter database 24. The implementation manners of the photographing element 21, the sensing element 22, and the processing device 23 can be the same as those of Figure 1 the photographing element 11, the sensing element 12, and the processing device 13, and will not be described herein again.

[0039] Figure 6 The parameter database 24 is shown as an external element of the processing device 23, and the parameter database 24 can also be provided in the processing device 23. The parameter database 24 can be implemented by a non-volatile memory, such as a read-only memory, a flash memory, and / or a non-volatile random access memory, etc., and the present invention is not limited thereto. The parameter database 24 corresponds to the electronic component and is connected to the processing device 23. The parameter database 24 can be used to store the first vibration parameter and the second vibration parameter of the electronic component.

[0040] To illustrate the operation mode of the vibration anomaly detection system 2 in more detail, please refer to Figure 6 and Figure 7 together, where Figure 7 is a flowchart of training a detection model shown in an embodiment of the present invention. Figure 7 The steps shown can be executed after Figure 3 the judgment result of step S107 in Figure 7 is "yes". As shown in

[0041]

[0042] In step S201, when the processing device 23 determines that both the first vibration parameter and the second vibration parameter meet their respective preset ranges, the processing device 23 can store the first vibration parameter and the second vibration parameter in the parameter database 24 corresponding to the electronic component. In step S203, the processing device 23 uses the parameter database 24 after adding the first vibration parameter and the second vibration parameter to train to generate a detection model. The detection model can be a discriminative artificial intelligence model, such as a support vector machine (SVM), a decision tree, a logistic regression, etc.In other words, the parameter database 24 may store multiple first vibration parameters that meet the preset range and multiple second vibration parameters that meet the preset range. The processing device 23 can use the multiple first vibration parameters and multiple second vibration parameters in the parameter database 24 for training to generate a detection model. And after the training of the detection model is completed, the processing device 23 can store the detection model. Accordingly, when the first vibration parameter and the second vibration parameter are obtained during subsequent detection of the electronic component, the processing device 23 can input the first vibration parameter and the second vibration parameter into the detection model to determine whether the electronic component is abnormal according to the detection result output by the detection model.

[0043] Please also refer to Figure 6 and Figure 8 , where Figure 8 is a flowchart for retraining the detection model shown according to an embodiment of the present invention. Figure 8 The steps shown can be executed after the judgment result of step S107 in Figure 3 is "no", and can be executed after step S109 in Figure 3 , before step S109, or executed simultaneously with step S109. As Figure 8 shown, the method for retraining the detection model may include: step S301: updating the parameter database with at least one parameter that does not meet the preset range; and step S303: retraining the detection model with the updated parameter database.

[0044] In step S301, when the processing device 23 determines according to the detection result output by the detection model that at least one of the first vibration parameter and the second vibration parameter does not meet their respective preset ranges, the processing device 23 can store the at least one parameter in the parameter database 24 corresponding to the electronic component to update the parameter database 24. In step S303, the processing device 23 retrains the detection model with the updated parameter database 24.

[0045] Taking the first vibration parameter not meeting the corresponding preset range as an example, the processing device 23 can mark the first vibration parameter as not meeting the preset range and store the marked first vibration parameter in the parameter database 24. Then, the processing device 23 can retrain the detection model with the updated parameter database 24. For example, for a product that has been repaired after leaving the factory, after detecting the electronic device through the vibration abnormality detection system and method of one or more of the above embodiments, if the detection result indicates that the first vibration parameter of the electronic device does not meet the preset range, the processing device 23 can mark the first vibration parameter as described above to retrain the detection model with the marked first vibration parameter. By adding the judgment criteria of defective products to the parameter database 24 as described above, as the data continues to increase, the judgment result of the detection model can be continuously optimized.

[0046] In addition, assuming that the first vibration parameter does not meet the corresponding preset range and the second vibration parameter meets the corresponding preset range, the processing device 23 can also mark the first vibration parameter as not meeting the preset range and mark the second vibration parameter as meeting the preset range, and store the marked first vibration parameter and the second vibration parameter in the parameter database 24 to retrain the detection model using the marked first vibration parameter and the second vibration parameter.

[0047] Through the vibration anomaly detection system and method of the above one or more embodiments, problematic electronic components can be identified and warned in real time and accurately, and can be used for online detection. Moreover, through the auxiliary detection of the vibration sensor, more comprehensive and accurate vibration information can be provided, which helps to more accurately determine whether there is an anomaly in the electronic component. In addition, through the parameter database of the above one or more embodiments, an accurate detection model can be established, and the judgment result of the detection model can be continuously optimized.

[0048] Although the present invention is disclosed above in the foregoing embodiments, it is not intended to limit the present invention. Any changes and modifications made without departing from the concept and scope of the present invention fall within the scope of patent protection of the present invention. For the scope of protection defined by the present invention, please refer to the claims.

Claims

1. A vibration anomaly detection method applicable to an electronic component, the method comprising performing, by a processing device: Obtaining multiple frames of detection images obtained by photographing the electronic component; Obtaining a first vibration parameter of the electronic component according to multiple blocks corresponding to the electronic component in the detection images; Obtaining a second vibration parameter generated by a vibration sensing component sensing the vibration of the electronic component; Judging whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges; and When it is judged that any one of the first vibration parameter and the second vibration parameter does not conform to its respective preset range, outputting a warning notification.

2. The vibration anomaly detection method according to claim 1, wherein judging whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges comprises: Inputting the first vibration parameter and the second vibration parameter into a detection model to obtain a detection result output by the detection model, wherein the detection result indicates whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges.

3. The vibration anomaly detection method according to claim 2, wherein the detection model is obtained by training with a parameter database corresponding to the electronic component, and the method further comprises: When the detection result indicates that at least one of the first vibration parameter and the second vibration parameter does not conform to its respective preset range, updating the parameter database with the at least one parameter; and Retraining the detection model with the updated parameter database.

4. The vibration anomaly detection method according to claim 1, further comprising: When it is judged that both the first vibration parameter and the second vibration parameter conform to their respective preset ranges, adding the first vibration parameter and the second vibration parameter to a parameter database corresponding to the electronic component; and Training with the parameter database to generate a detection model.

5. The vibration anomaly detection method according to claim 1, wherein each of the first vibration parameter and the second vibration parameter comprises at least one of frequency, displacement and amplitude.

6. A vibration anomaly detection system applicable to an electronic component, comprising: A photographing component for photographing the electronic component to generate multiple frames of detection images; A sensing component for sensing the vibration of the electronic component; and A processing device connected to the photographing component and the sensing component, the processing device being configured to obtain a first vibration parameter according to multiple blocks corresponding to the electronic component in the detection images, obtain a second vibration parameter from the sensing component, judge whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges, and output a warning notification when it is judged that any one of the first vibration parameter and the second vibration parameter does not conform to its respective preset range.

7. The vibration anomaly detection system according to claim 6, wherein the processing device is further configured to store a detection model, and the processing device inputs the first vibration parameter and the second vibration parameter into the detection model to obtain a detection result output by the detection model, wherein the detection result indicates whether the first vibration parameter and the second vibration parameter conform to their respective preset ranges.

8. The vibration abnormality detection system according to claim 7 further includes; A parameter database corresponding to the electronic component and connected to the processing device, wherein the processing device is trained with the parameter database to generate the detection model, and the processing device is further configured to update the parameter database with the at least one parameter and retrain the detection model with the updated parameter database when the detection result indicates that at least one of the first vibration parameter and the second vibration parameter does not conform to the respective preset range.

9. The vibration abnormality detection system according to claim 6 further includes; A parameter database corresponding to the electronic component and connected to the processing device, wherein the processing device is further configured to add the first vibration parameter and the second vibration parameter to the parameter database and train with the parameter database to generate a detection model when it is determined that both the first vibration parameter and the second vibration parameter conform to the respective preset ranges.

10. In the vibration abnormality detection system according to claim 6, wherein each of the first vibration parameter and the second vibration parameter includes at least one of frequency, displacement, and amplitude.

Images