Product quality evaluation and fault detection method based on acoustic characteristics

Abstract

The invention provides a product quality evaluation and fault detection method based on acoustic characteristics, and the method comprises the following steps: S1, fixing a device on the surface of a detected device, and connecting the device with a local server in a wireless manner; s2, automatically collecting a sound file generated when the equipment runs, and sending the sound file to a local server for analysis; s3, carrying out time domain analysis, identifying a circulation point by using an amplitude envelope line, and carrying out classification processing on the signals; s4, scoring the preprocessed cyclic sound one by one according to a psychoacoustic evaluation system, and outputting a total score of equipment audio; s5, comparing single cycles, and identifying impact features by using the spectrogram after Fourier transform; and S6, comparing a result with a local template, and automatically generating a detection report. According to the method, sound evaluation datamation is realized, the test efficiency is improved, and abnormal sound can be monitored to early warn potential problems in advance. The device adopts wireless communication, supports simultaneous acquisition and comparison of multiple devices, reduces the arrangement difficulty of the sensors, and is not influenced by movement of the devices.

Description

Technical Field

[0001] The present invention belongs to the technical field of fault diagnosis, and in particular relates to a product quality assessment and fault detection method based on acoustic features. Background Art

[0002] At present, fault diagnosis technology generally relies on means such as vibration analysis or audio analysis. Although vibration analysis is effective, it requires direct contact with the surface of the equipment, so its application is limited in harsh environments such as high temperature and high corrosion. In contrast, although audio analysis avoids the limitation of direct contact, its implementation often relies on complex test layouts, requiring advance knowledge of the equipment speed and analysis of the status of components corresponding to specific frequency bands. This process is not only costly and cumbersome, but also easily interfered by environmental noise, resulting in inaccurate diagnostic results.

[0003] During long-term operation, equipment in the production environment will inevitably face problems such as aging and wear of parts, which will cause unpredictable failures. Once these failures occur, if the manufacturer does not respond in time, it is very easy to cause production line interruptions and cause huge economic losses. Given that traditional fault diagnosis methods are powerless in the face of complex and changing production environments, there is an urgent need for a more flexible and efficient fault warning mechanism to achieve early detection and rapid response to potential faults in production equipment, thereby improving maintenance efficiency and ensuring production continuity. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a product quality assessment and fault detection method based on acoustic features, so as to realize the monitoring of the status of production equipment and the early warning and diagnosis of potential faults with a more flexible and efficient fault warning mechanism, so as to solve the limitations of the prior art, such as installation environment restrictions, high layout costs and cumbersome operations.

[0005] Technical solution: A product quality assessment and fault detection method based on acoustic features of the present invention comprises the following steps:

[0006] S1: device fixing and parameter setting step, wherein the device has one or more acoustic sensors, which are used to be fixed on the surface of the device under test and establish a connection with a local server through a wireless communication module; after setting the relevant parameters, a data transmission link is established with the local server through wireless transmission;

[0007] S2: sound file collection step, the device automatically collects the sound files generated by the device under test during operation by means of external signal triggering or timed automatic triggering, and sends these sound files to the local server for analysis in real time or timed manner;

[0008] S3: audio signal processing step, performing time domain analysis on the received original audio signal on the local server, identifying each loop point using the amplitude envelope technique, and determining the start and end points of each loop of the audio signal;

[0009] S4: Audio scoring step: scoring the pre-processed single loop sounds one by one according to the psychoacoustic evaluation system. The scoring process takes into account the psychoacoustic parameters of the sound and outputs the total audio score of the device according to the preset scoring model as the basis for product quality evaluation;

[0010] S5: Fault feature identification step, comparing the pre-processed single cycle sounds one by one, using Fourier transform technology to generate a spectrum diagram, and looking for impact features or other abnormal frequency distributions appearing in the cycle in the spectrum diagram to identify possible fault signals;

[0011] S6: Test report generation and early warning step, compare and analyze the results of steps S4 and S5 with the template results stored on the local server, and automatically generate a detailed test report containing information such as the device audio quality score, fault feature description and location; when the test result does not exceed the set threshold, the green light on the device surface is always on; once the test result exceeds the set threshold, an instruction is sent to the device through the wireless communication module, triggering the red light on the device surface to light up, thereby realizing early warning of the fault.

[0012] Further, in step S1, the device includes:

[0013] A wireless communication module, used for data transmission with a local server;

[0014] A data processing unit for performing computational tasks such as preprocessing of audio signals, time domain analysis, psychoacoustic scoring, and spectrum analysis;

[0015] Warning devices, such as red lights or other forms of visual or audible alarms, are used to issue early warning signals when an abnormality is detected;

[0016] The user interaction interface optionally provides an interface for setting parameters, viewing test reports, and receiving warning information.

[0017] Furthermore, the device can adapt to different installation surface conditions, and can automatically collect the sound generated during product operation according to preset rules, and send the data to a local server for analysis via wireless transmission. Inspectors can not only visually determine whether there are abnormal conditions through the signal lights on the device, but also view detailed analysis reports through networked devices, USB storage devices or other data transmission methods.

[0018] Furthermore, in step S3, the process of identifying each cycle point using the amplitude envelope technology includes: extracting the absolute value of the audio signal to obtain a curve representing the change in signal strength; then, smoothing the signal based on filtering processing to obtain a clearer envelope curve; finally, identifying the peaks and troughs through a preset amplitude difference threshold, thereby determining the starting point and end point of each cycle of the audio signal.

[0019] Furthermore, in step S4, the psychoacoustic parameters of the sound include the pitch, loudness and sharpness of the sound.

[0020] Furthermore, in step S4, the audio signal is decomposed into sine waves and cosine waves of different frequencies by Fourier transform technology to construct a spectrum diagram. The spectrum diagram is analyzed to search for instantaneous cross-band features and determine whether there is an impact sound in the cycle. At the same time, the frequency band characteristic curve is compared with the historical model to identify abnormal ripples in a specific frequency band, thereby detecting equipment failure. This method can monitor the equipment status in real time and improve detection sensitivity and accuracy.

[0021] Furthermore, the model allows customers to adjust model parameters according to different product categories and testing requirements to improve the accuracy and pertinence of the scoring.

[0022] Preferably, the fixing method includes fixing methods such as liquid glue, solid glue or bolts, which can be selected according to the actual surface state of the product to be measured to ensure the stability of the sensor.

[0023] Preferably, the external signal trigger can set the timing of the signal trigger according to the actual working conditions, ensuring that the collected sound is more in line with the actual situation without affecting the normal operation of the equipment.

[0024] Preferably, the audio signal processing can reduce the impact of the environment based on a specific filtering algorithm. For products with periodic motion, the loop points can be identified through the amplitude envelope technology to provide a basis for subsequent judgment.

[0025] Preferably, the audio score may be based on an evaluation method of conventional psychoacoustic parameters and impact sound calculation, and may be adjusted according to actual products to improve the rationality of the score.

[0026] Preferably, the fault feature identification includes comparing historical data with the cyclic segmentation based on S3, and judging whether the sound is a fault according to the difference points, so as to ensure the accuracy of identification.

[0027] Preferably, the early warning function includes a light prompt on the surface of the device and a display of the time when the abnormality occurred and the historical change record through an interactive interface to enhance the user experience.

[0028] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: accurately identifying the cycle segments through envelope analysis and screening out the same features of the same cycle, effectively avoiding the influence of environmental noise on the analysis results; locating the fault location by using characteristic frequency band ripple analysis, improving the accuracy of fault detection; customers can improve diagnostic efficiency by accumulating feature libraries, reduce dependence on manual experience, and realize automation and intelligence of fault diagnosis. In addition, the present invention can determine the fixed position and method according to the surface state of the product under test, and directly score through the recording file without presetting the relevant motion indicators of the product under test. The score obtained based on the psychological evaluation parameters can adjust the evaluation coefficient according to the actual product. In addition to being able to diagnose faults, it can also evaluate the product sound quality. Finally, the fault state of the equipment can be intuitively judged by the indicator light on the surface of the equipment, and the fault can be identified by observation, providing an intuitive and efficient fault detection method. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flowchart of a method and device for product quality assessment and fault detection based on acoustic features;

[0030] Figure 2 A schematic diagram of a scenario of the method of the present invention;

[0031] Figure 3 It is the installation schematic diagram of the method of the present invention (front view);

[0032] Figure 4 It is a schematic diagram of the planar installation of the method of the present invention (side view);

[0033] Figure 5 Schematic diagram of curved surface installation of the method of the present invention (side view);

[0034] Figure 6 Flowchart for establishing a psychological scoring model. DETAILED DESCRIPTION

[0035] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0036] Combine the following Figures 1 to 6 The method and device for product quality assessment and fault detection based on acoustic features according to an embodiment of the present invention include the following steps:

[0037] S1: Device fixing and parameter setting steps, refer to Figure 2 , Figure 4 and Figure 5 The fault diagnosis device has one or more acoustic sensors, which are used to be fixed on the surface of the device under test. Depending on the surface conditions, liquid, solid or threaded fixing methods can be selected. Press Figure 3After the button marked as 3 is powered on, a connection is established with the local server based on the wireless communication module. After setting the evaluation coefficient, sampling frequency and other parameters, a stable data transmission link is established with the local server through wireless transmission;

[0038] S2: Sound file collection steps, refer to Figure 2 The device automatically collects the sound files generated by the device under test during operation, and sends these sound files to the local server in real time or at a fixed time for further analysis;

[0039] S3: audio signal processing step, performing time domain analysis on the received original audio signal on the local server, specifically using amplitude envelope technology to identify each loop point, and classifying the signal for subsequent analysis;

[0040] S4: Audio scoring step: According to the psychoacoustic evaluation system, the pre-processed single loop sound is scored one by one. The scoring process takes into account the psychoacoustic parameters such as the pitch, loudness, and sharpness of the sound, and outputs the total audio score of the device according to the preset scoring model as the basis for product quality evaluation;

[0041] S5: Fault feature identification step, comparing the pre-processed single cycle sounds one by one, using Fourier transform technology to generate a spectrum diagram, and looking for impact features or other abnormal frequency distributions appearing in the cycle in the spectrum diagram to identify possible fault signals;

[0042] S6: Test report generation and early warning step, compare and analyze the results of steps S4 and S5 with the template results stored on the local server, and automatically generate a detailed test report containing information such as the device audio quality score, fault feature description and location. Figure 3 The indicator lights marked 4 and 5 are always green when the detection result does not exceed the set threshold. Once the detection result exceeds the set threshold, a command is sent to the device through the wireless communication module, triggering the red light on the surface of the device to light up, thus realizing early warning of faults.

[0043] The method and device for product quality assessment and fault detection based on acoustic features can first determine the fixed position and method according to the surface state of the tested product; secondly, connect to the local server for the surrounding environment; and then directly score through the recording file without presetting the relevant motion indicators of the tested product. In addition, the score obtained based on the psychological evaluation parameters can adjust the evaluation coefficient according to the actual product. The specific process of establishing the scoring model can be referred to Figure 6This method can not only diagnose faults, but also evaluate the sound quality of the product; finally, through the indicator lights on the surface of the device, users can intuitively judge the fault status of the device, thereby facilitating a quick response.

[0044] It should be noted that, in the present invention, the use of "includes", "comprising" or any other variant words is intended to express a non-exclusive meaning, so that a process, method, article or terminal device including a series of elements includes not only these explicitly listed elements, but also includes elements that are not explicitly listed, as well as elements inherent to the process, method, article or terminal device. Therefore, without adding more restrictions, the use of the elements specified by the sentence "includes..." or "comprising..." does not exclude the possibility of other elements in the process, method, article or terminal device. In addition, in this article, the words "greater than", "less than", "exceed" and the like are understood to exclude the number itself; while the words "above", "below", "within" and the like are understood to include the number itself.

[0045] Although the above embodiments have been described, once those skilled in the art have mastered the basic innovative ideas of the present invention, other modifications and changes may be made to these embodiments. Therefore, the above embodiments are only used to explain the present invention and should not limit the scope of patent protection of the present invention. Without exceeding the scope of the present invention, any equivalent structure or process transformation using the contents of the present specification and drawings, or innovation directly or indirectly applied in other related technical fields, are included in the scope of patent protection of the present invention.

Claims

1. A product quality assessment and fault detection method based on acoustic features, characterized in that: The steps include: S1: device fixing and parameter setting step, wherein the device has one or more acoustic sensors, which are used to be fixed on the surface of the device under test and establish a connection with a local server through a wireless communication module; after setting the relevant parameters, a data transmission link is established with the local server through wireless transmission; S2: sound file collection step, the device automatically collects the sound files generated by the device under test during operation by means of external signal triggering or timed automatic triggering, and sends these sound files to the local server for analysis in real time or timed manner; S3: audio signal processing step, performing time domain analysis on the received original audio signal on the local server, identifying each loop point using the amplitude envelope technique, and determining the start and end points of each loop of the audio signal; S4: Audio scoring step: scoring the pre-processed single loop sounds one by one according to the psychoacoustic evaluation system. The scoring process takes into account the psychoacoustic parameters of the sound and outputs the total audio score of the device according to the preset scoring model as the basis for product quality evaluation; S5: Fault feature identification step, comparing the pre-processed single cycle sounds one by one, using Fourier transform technology to generate a spectrum diagram, and looking for impact features or other abnormal frequency distributions appearing in the cycle in the spectrum diagram to identify possible fault signals; S6: Test report generation and early warning step, compare and analyze the results of steps S4 and S5 with the template results stored on the local server, and automatically generate a detailed test report containing information such as the device audio quality score, fault feature description and location; when the test result does not exceed the set threshold, the green light on the device surface is always on; once the test result exceeds the set threshold, an instruction is sent to the device through the wireless communication module, triggering the red light on the device surface to light up, thereby realizing early warning of the fault.

2. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: In step S1, the device comprises: Wireless communication module, used for data transmission with a local server; A data processing unit for performing computational tasks such as preprocessing of audio signals, time domain analysis, psychoacoustic scoring, and spectrum analysis; Warning devices, such as red lights or other forms of visual or audible alarms, are used to issue early warning signals when an abnormality is detected; The user interaction interface optionally provides an interface for setting parameters, viewing test reports, and receiving warning information.

3. The method for product quality assessment and fault detection based on acoustic features according to claim 2, characterized in that: The device can adapt to different installation surface conditions, automatically collect the sound generated by the product during operation according to preset rules, and send the data to the local server for analysis via wireless transmission. Inspection personnel can not only visually judge whether there is an abnormality through the signal light on the device, but also view detailed analysis reports through networked devices, USB storage devices or other data transmission methods.

4. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: In step S3, the process of identifying each cycle point using the amplitude envelope technology includes: extracting the absolute value of the audio signal to obtain a curve representing the change in signal strength; then, smoothing the signal based on filtering processing to obtain a clearer envelope curve; finally, identifying the peaks and troughs through a preset amplitude difference threshold, thereby determining the starting point and end point of each cycle of the audio signal.

5. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: In step S4, the psychoacoustic parameters of the sound include the pitch, loudness and sharpness of the sound.

6. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: In step S4, the Fourier transform technology is used to generate a spectrum diagram, and the impact characteristics or other abnormal frequency distributions appearing in the cycle are searched in the spectrum diagram. Specifically, the audio signal is decomposed into sine waves and cosine waves of different frequencies, a spectrum diagram is constructed, the spectrum diagram is analyzed, and instantaneous cross-band characteristics are searched to determine whether there is an impact sound in the cycle; at the same time, the frequency band characteristic curve is compared with the historical model to identify abnormal ripples in a specific frequency band, thereby detecting equipment failure.

7. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: The model allows customers to adjust model parameters according to different product categories and testing requirements to improve the accuracy and pertinence of the scoring.

8. The method for product quality assessment and fault detection based on acoustic features according to claim 1, characterized in that: The device can be fixed in a liquid glue, solid glue or bolt manner according to the actual surface state of the product being tested.

Images