Acoustic diagnosis device and acoustic diagnosis method

The acoustic diagnostic device uses sound and image sensors with AI to analyze sound pressure levels and identify abnormal parts in complex equipment, overcoming the challenge of varying noise timing and frequency, ensuring precise detection and improved maintenance.

WO2024242153A1PCT designated stage expired Publication Date: 2024-11-28JFE STEEL CORP

Patent Information

Application Number
PCT/JP2024/018926
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-24
Filing Date
2024-05-23
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Existing acoustic diagnostic devices struggle to accurately identify abnormal parts in equipment composed of multiple components, as the timing and frequency of operating noise vary by part, making it difficult to determine the source of abnormal sounds.

Method used

An acoustic diagnostic device that collects operation sounds and images using sensors and AI models to extract sound pressure levels for specific frequency bands, calculates deviation degrees, and identifies abnormal regions based on these data, considering operating conditions.

Benefits of technology

Enables accurate identification of abnormal parts by analyzing sound pressure levels across operating sections and frequency bands, improving equipment maintenance and product quality by quickly pinpointing issues.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024018926_28112024_PF_FP_ABST
    Figure JP2024018926_28112024_PF_FP_ABST
Patent Text Reader

Abstract

An acoustic diagnosis device according to the present invention comprises: an operation sound collection unit that collects operation sound of equipment; an operation section operation sound extraction unit that collects images indicating the operation state of the equipment, detects predetermined operation sections of the equipment using the collected images, and extracts operation sound for each of the detected operation sections from the operation sound collected by the operation sound collection unit; a sound pressure level acquisition unit that acquires a sound pressure level for each frequency band predetermined for each operation section with respect to the operation sound extracted by the operation section operation sound extraction unit for each operation section; a deviation degree computation unit that computes the degree of deviation from a sound pressure level at the time of normal operation of the same equipment, operation section, and frequency band with respect to the sound pressure level acquired by the sound pressure level acquisition unit for each operation section and each frequency band; and an abnormal site identification unit that identifies an abnormal site of the equipment on the basis of the degree of deviation computed by the deviation degree computation unit.
Need to check novelty before this filing date? Find Prior Art

Description

Acoustic diagnostic device and acoustic diagnostic method

[0001] The present invention relates to an acoustic diagnostic device and an acoustic diagnostic method for identifying an abnormal part of equipment using an acoustic signal generated during operation of the equipment.

[0002] Patent Document 1 describes a diagnostic device that diagnoses an object to be diagnosed by comparing an observed waveform observed from the object to be diagnosed with a reference waveform. Patent Document 2 describes a state estimation device that estimates a change in the state of equipment based on waveforms and frequency information detected by multiple sensors installed in the equipment. Patent Document 3 describes a diagnostic device that detects chattering of rolling mill rolls using the frequency of an acoustic signal detected by an acoustic sensor installed in a rolling mill.

[0003] JP 2008-164490 A International Publication No. 2019 / 017345 International Publication No. 2020 / 157818

[0004] The devices described in Patent Documents 1 to 3 may not be able to accurately identify an abnormal part of equipment, particularly in equipment that is composed of multiple parts. Specifically, when equipment is composed of multiple parts such as bearings, reducers, cranks, dies, rolls, and blocks, the timing at which operating sounds are generated varies depending on the parts. Therefore, unless the timing and frequency of the abnormal sound are taken into consideration, it may not be possible to determine which part the abnormal sound is coming from, and the abnormal part of the equipment may not be able to be accurately identified.

[0005] The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to provide an acoustic diagnostic device and an acoustic diagnostic method that can accurately identify an abnormal portion of equipment.

[0006] The acoustic diagnostic device of the present invention is an acoustic diagnostic device that identifies abnormal parts of equipment using acoustic signals generated when the equipment is in operation, and includes: an operation sound collection unit that collects operation sounds of the equipment; an operation section operation sound extraction unit that collects images showing the operating state of the equipment, detects predetermined operation sections of the equipment using the collected images, and extracts the operation sounds for each operation section detected from the operation sounds collected by the operation sound collection unit; a sound pressure level acquisition unit that acquires sound pressure levels for each frequency band predetermined for each operation section for the operation sounds extracted by the operation section operation sound extraction unit; a deviation calculation unit that calculates the degree of deviation from the sound pressure level during normal operation of the same equipment, operation section, and frequency band for the sound pressure levels for each operation section and each frequency band acquired by the sound pressure level acquisition unit; and an abnormal part identification unit that identifies abnormal parts of the equipment based on the degree of deviation calculated by the deviation calculation unit.

[0007] The operation section operation sound extraction unit may input an image showing the operation state of the equipment, acquired using an image sensor, into an artificial intelligence model that uses an image showing the operation state of the equipment as input data and the operation state of the equipment shown in the image as output data, and thereby detect a predetermined operation section of the equipment using the image showing the operation state of the equipment.

[0008] The sound pressure level acquisition unit classifies each operating sound based on predetermined operating conditions of the equipment and acquires the sound pressure level, and the deviation calculation unit calculates the degree of deviation from the sound pressure level during normal operation of the same equipment, operating section, frequency band, and operating condition, for the sound pressure level for each operating section, frequency band, and operating condition acquired by the sound pressure level acquisition unit.

[0009] The system may include a database unit that stores conditions for extracting the operating sounds of the equipment and data necessary for identifying abnormal parts of the equipment, and the abnormal part identification unit may identify abnormal parts of the equipment based on a pattern of deviation degrees that is based on the deviation degrees in each frequency band obtained for each operating section stored in the database unit.

[0010] It is preferable to provide a normal state sound pressure level acquisition unit that acquires sound pressure levels for each frequency band predetermined for each operation section for the operation sound extracted by the operation section operation sound extraction unit, and a normal distribution determination and recording unit that classifies the sound pressure levels acquired by the normal state sound pressure level acquisition unit into those for each equipment, each operation section, and each frequency band, and records them as sound pressure levels during normal operation of the equipment.

[0011] The normal distribution determination and recording unit may record, for each piece of equipment, a table containing information indicating the relationship between a pattern of deviations of sound pressure levels from the sound pressure level during normal operation in each frequency band determined for each operating section and abnormal parts of the equipment.

[0012] The normal distribution determination and recording unit may record a table including information indicating the operation interval, the pattern of the deviation, and the relationship between the operating conditions of the equipment and the abnormal portion of the equipment.

[0013] The acoustic diagnostic method according to the present invention is an acoustic diagnostic method for identifying an abnormal part of equipment using an acoustic signal generated when the equipment is in operation, and includes: an operation sound collection step of collecting operation sounds of the equipment; an operation section operation sound extraction step of collecting images showing the operating state of the equipment, detecting predetermined operation sections of the equipment using the collected images, and extracting the operation sounds for each operation section detected from the operation sounds collected in the operation sound collection step; a sound pressure level acquisition step of acquiring sound pressure levels for each frequency band predetermined for each operation section for the operation sounds extracted in the operation section operation sound extraction step; a deviation calculation step of calculating a degree of deviation from a sound pressure level during normal operation of the same equipment, operation section, and frequency band for the sound pressure levels for each operation section and each frequency band acquired in the sound pressure level acquisition step; and an abnormal part identification step of identifying an abnormal part of the equipment based on the degree of deviation calculated in the deviation calculation step.

[0014] It is preferable that the method further includes a normal state sound pressure level acquisition step of acquiring a sound pressure level for each predetermined frequency band for each operation section for the operation sound extracted in the operation section operation sound extraction step, and a normal distribution determination and recording step of classifying the sound pressure levels acquired in the normal state sound pressure level acquisition step into those for each equipment, each operation section, and each frequency band, and recording them as sound pressure levels during normal operation of the equipment.

[0015] The operation section operation sound extraction step may include a step of detecting a predetermined operation section of the equipment using the image showing the operation state of the equipment by inputting an image showing the operation state of the equipment acquired using an image sensor into an artificial intelligence model that uses an image showing the operation state of the equipment as input data and the operation state of the equipment shown in the image as output data.

[0016] According to the acoustic diagnostic device and acoustic diagnostic method of the present invention, abnormal parts of equipment are identified based on sound pressure levels for each operating section and each frequency band of the equipment, so that abnormal parts of the equipment can be identified with high accuracy.

[0017] FIG. 1 is a block diagram showing the configuration of an acoustic diagnostic apparatus according to one embodiment of the present invention. FIG. 2 is a flowchart showing the flow of data recording processing according to one embodiment of the present invention. FIG. 3 is a diagram for explaining data recording processing according to one embodiment of the present invention. FIG. 4 is a diagram for explaining data recording processing according to one embodiment of the present invention. FIG. 5 is a diagram for explaining data recording processing according to one embodiment of the present invention. FIG. 6 is a flowchart showing the flow of abnormal portion identification processing according to one embodiment of the present invention. FIG. 7 is a diagram for explaining a specific example of abnormal portion identification processing according to one embodiment of the present invention. FIG. 8 is a diagram for explaining abnormal portion identification processing of an example. FIG. 9 is a diagram showing the frequency distribution of sound pressure levels when the operating conditions of the press equipment are limited and when they are not limited.

[0018] An acoustic diagnostic apparatus and an acoustic diagnostic method according to an embodiment of the present invention will be described below with reference to the drawings.

[0019] [Configuration] Fig. 1 is a block diagram showing the configuration of an acoustic diagnostic device according to one embodiment of the present invention. As shown in Fig. 1, the acoustic diagnostic device 1 according to one embodiment of the present invention is a device that identifies an abnormal portion of one or more predetermined pieces of equipment to be diagnosed (hereinafter abbreviated as equipment) A using acoustic signals generated during operation of the equipment A, and is configured using an information processing device such as a computer. Here, the equipment A is equipment composed of a plurality of parts, and examples of such equipment include steel-making equipment such as press equipment and rolling equipment.

[0020] The acoustic diagnostic device 1 includes an operation sound collection unit 11, an operation section operation sound extraction unit 12, a normal state sound pressure level acquisition unit 13, a normal distribution determination and recording unit 14, a sound pressure level acquisition unit 15, a deviation degree calculation unit 16, an abnormal part identification unit 17, and a database unit 18. The operation sound collection unit 11, the operation section operation sound extraction unit 12, the normal state sound pressure level acquisition unit 13, the normal distribution determination and recording unit 14, the sound pressure level acquisition unit 15, the deviation degree calculation unit 16, and the abnormal part identification unit 17 are functional blocks realized by an arithmetic processing unit such as a CPU inside the information processing device executing a computer program. The functions of each unit will be described later.

[0021] The database unit 18 is configured by a storage device such as an HDD that constitutes an information processing device, and includes a cutting condition database 18a, a frequency band database 18b, a normal distribution database 18c, an operating condition database 18d, and a deviation pattern database 18e. Details of each database will be described later. Note that the database unit 18 may be located outside the acoustic diagnostic device 1, and the acoustic diagnostic device 1 may be configured to acquire data from the database unit 18 via a telecommunications line.

[0022] The acoustic diagnostic device 1 having such a configuration executes the data recording process and the abnormal part identification process described below to identify the abnormal part of the equipment A. The operation of the acoustic diagnostic device 1 when executing the data recording process and the abnormal part identification process will be described below.

[0023] 2 is a flowchart showing the flow of the data recording process according to one embodiment of the present invention. The flowchart shown in FIG. 2 starts when an execution command for the data recording process is input to the acoustic diagnostic device 1, and the data recording process proceeds to step S1.

[0024] In the processing of step S1, the operation sound collection unit 11 collects sound pressure signals of the operation sound of equipment A during normal operation at predetermined time intervals using j types (j≧1) of acoustic sensors 20, such as microphones, installed near equipment A. In addition, the operation section operation sound extraction unit 12 collects images of equipment A during normal operation using an image sensor 21, such as a camera, installed near equipment A. For example, in the example shown in FIG. 3( a), when a workpiece is transported to the equipment by a transport machine and processed therein, the operation sound collection unit 11 collects the operation sounds of the equipment and the transport machine, and the operation section operation sound extraction unit 12 collects images of the equipment including the workpiece and the transport machine. This completes the processing of step S1, and the data recording processing proceeds to the processing of step S2.

[0025] In the processing of step S2, the operation interval operation sound extraction unit 12 detects the start time and end time of k types (k≧1) of operation intervals of the equipment A determined in advance using images of the equipment A during normal operation collected in the processing of step S1 (operation interval detection processing). For example, in the example shown in FIG. 3(b), the operation interval operation sound extraction unit 12 detects the time when a workpiece enters the equipment using images collected by the image sensor 21. Details of this operation interval detection processing will be described later with reference to FIGS. 4 and 5. Then, the operation interval operation sound extraction unit 12 extracts a sound pressure signal for each operation interval from the sound pressure signals collected in the processing of step S1 based on the start time and end time of the detected operation interval. For example, in the example shown in FIG. 3(c), the operation interval operation sound extraction unit 12 extracts a sound pressure signal for each operation interval from the sound pressure signal collected in the processing of step S1 based on the time when the workpiece enters the equipment and the time when the workpiece leaves the equipment. 1 The sound pressure signals are extracted from operation section 1, which is a section of one second, and operation section 2, which is a section of a predetermined time from the time the workpiece leaves the equipment. This completes the processing of step S2, and the data recording processing proceeds to processing of step S3.

[0026] In the process of step S3, first, the normal state sound pressure level acquisition unit 13 converts the sound pressure signal for each operation section extracted in the process of step S2 into a sound pressure level for each frequency using a Fourier transform, an octave band filter, or the like. Next, the normal state sound pressure level acquisition unit 13 reads information (such as minimum and maximum frequency values) related to l types (l≧1) of frequency bands used for abnormality detection predetermined for each operation section from the frequency band database 18b. Then, the normal state sound pressure level acquisition unit 13 acquires the sound pressure level for each l types of frequency bands (sound pressure level of normal operation sound) from the converted sound pressure level data using a bandpass filter, for example. For example, in the example shown in FIG. 3(d), the normal state sound pressure level acquisition unit 13 acquires the sound pressure level of normal operation sound for each of predetermined frequency bands 1 to 3. This completes the process of step S3, and the data recording process proceeds to the process of step S4.

[0027] In the process of step S4, the normal distribution determination and recording unit 14 classifies the sound pressure levels of the normal operation sounds acquired in the process of step S3 into operation sections and frequency bands, and records data on the sound pressure level distribution of the normal operation sounds and their feature quantities (average value, standard deviation, etc.) in the normal distribution database 18c. This completes the process of step S4, and the series of data recording processes ends.

[0028] [Operation Section Detection Process] In the operation section detection process in the processing of step S2 above, the operation section operation sound extraction unit 12 determines the start time and end time of k types of operation sections of predetermined equipment A using an artificial intelligence model trained using images collected by the image sensor 21. Here, the artificial intelligence model is a learning model in which the images collected by the image sensor 21 are used as input data and the operation state of equipment A indicated by the images is used as output data. The artificial intelligence model is generated by training a set of performance data of the images collected by the image sensor 21 and data on the operation state of equipment A indicated by the images as training data. The artificial intelligence model may be any model that can be used for image judgment such as deep learning or machine learning.

[0029] For example, when determining the time when a workpiece enters the equipment (workpiece entry detection) and the time when the workpiece leaves the equipment (workpiece exit detection) using images collected by the image sensor 21, an image in which the workpiece and the equipment are captured in a single image is selected as the image to be collected by the image sensor 21. Examples of workpiece entry detection images (images 1 and 2) and workpiece exit detection images (images 3 and 4) are shown in Figures 4(a) and 4(b). As shown in Figure 4(a), images of the moment when the workpiece enters the equipment and the state of the workpiece entering and moving forward in the equipment are also used to learn the time when the workpiece entered the equipment. Furthermore, as shown in Figure 4(b), images of the moment when the workpiece leaves the equipment and the state of the workpiece leaving the equipment and moving forward in the equipment are also used to learn the time when the workpiece left the equipment.

[0030] An example of the processing flow of the operation section operation sound extraction unit 12 from image acquisition of equipment A to workpiece entry and exit detection is shown in Figure 5. As shown in Figure 5, the operation section operation sound extraction unit 12 acquires an image using the image sensor 21 and inputs the acquired image data to the artificial intelligence model (step S21). If the artificial intelligence model outputs an operation state in which a workpiece has entered the equipment (step S22: Yes), the operation section operation sound extraction unit 12 turns the trigger on (step S23). On the other hand, if the artificial intelligence model does not output an operation state in which a workpiece has entered the equipment (step S22: No), the operation section operation sound extraction unit 12 turns the trigger off (step S24) and executes the processing of step S21 again.

[0031] If the trigger is turned on in the processing of step S23, the operation section operation sound extraction unit 12 acquires an image using the image sensor 21 and inputs the acquired image data to the artificial intelligence model (step S25). Then, if the artificial intelligence model outputs an operation state in which a workpiece has left the equipment (step S26: Yes), the operation section operation sound extraction unit 12 sets the trigger to an off state (step S27). This makes it possible to detect the time when the workpiece entered equipment A and the time when the workpiece left equipment A based on the time when the trigger became on and off. On the other hand, if the artificial intelligence model does not output an operation state in which a workpiece has left the equipment (step S26: No), the operation section operation sound extraction unit 12 executes the processing of step S25 again.

[0032] [Abnormality Region Identification Processing] Fig. 6 is a flowchart showing the flow of abnormality region identification processing according to one embodiment of the present invention. The flowchart shown in Fig. 6 starts when an execution command for the abnormality region identification processing is input to the acoustic diagnostic device 1, and the abnormality region identification processing proceeds to step S11.

[0033] In the processing of step S11, the operation sound collection unit 11 uses the acoustic sensor 20 installed near the equipment A to collect sound pressure signals of the operation sounds of the equipment A during a predetermined period including k types (k≧1) of operation intervals of the pre-classified equipment A at predetermined time intervals. In addition, the operation interval operation sound extraction unit 12 uses the image sensor 21 to collect images of the equipment A during normal operation. This completes the processing of step S11, and the abnormal part identification processing proceeds to the processing of step S12.

[0034] In the processing of step S12, the operation interval operation sound extraction unit 12 detects the start time and end time of k types (k≧1) of operation intervals of predetermined equipment A using images of equipment A in normal operation collected in the processing of step S11. Then, based on the start time and end time of the detected operation intervals, the operation interval operation sound extraction unit 12 extracts sound pressure signals for each of the k types of operation intervals from the sound pressure signals collected in the processing of step S11. The processing flow of this step S12 is the same as the processing flow of step S2 described above, so a detailed description thereof will be omitted. This completes the processing of step S12, and the abnormal part identification processing proceeds to the processing of step S13.

[0035] In the processing of step S13, the sound pressure level acquisition unit 15 acquires sound pressure levels for each of k types of operating sections and l types of frequency bands from the sound pressure signal of the operation sound extracted in the processing of step S12. The processing flow of this step S13 is the same as the processing flow of step S3 described above, so a detailed description thereof will be omitted. Note that the sound pressure level acquisition unit 15 may read information on m types of operating conditions that affect the operation of the equipment A from the operating condition database 18d, and acquire sound pressure levels for each of k types of operating sections, l types of frequency bands, and m types of operating conditions. This completes the processing of step S13, and the abnormal part identification processing proceeds to the processing of step S14.

[0036] In the processing of step S14, the deviation calculation unit 16 reads information on the sound pressure level distribution during normal operation and its feature quantities from the normal distribution database 18c for each of k types of operation intervals and l types of frequency bands of equipment A. Next, using the read information, the deviation calculation unit 16 calculates two or more deviations of the sound pressure level of the operation sound acquired in the processing of step S13 relative to the sound pressure level during normal operation for each of k types of operation intervals and l types of frequency bands. Then, using the calculated deviations, the deviation calculation unit 16 determines whether equipment A is normal (0) or abnormal (1) for each deviation, and generates an array pattern vector of the determination results (0, 1) for each operation interval as a deviation pattern.

[0037] In the processing of step S13, when m types of operating conditions are taken into consideration, the deviation calculation unit 16 calculates the deviation by referring to a data table such as that shown in Table 1 below. The data table shown in Table 1 below shows information on the sound pressure level distribution and its feature quantities during normal operation for each of k types of operating sections, l type of frequency bands, and m types of operating conditions. The data table shown in Table 1 below corresponds to the contents of the extraction condition database 18a, frequency band database 18b, normal distribution database 18c, and operating condition database 18d. This completes the processing of step S14, and the abnormal part identification processing proceeds to the processing of step S15.

[0038]

[0039] In the process of step S15, the abnormal part identifying unit 17 identifies an abnormal part of the equipment A based on the deviation pattern y' generated in the process of step S14. Specifically, as shown in FIGS. 7(a) and 7(b), the abnormal part identifying unit 17 identifies an abnormal part of the equipment A based on the operation section and deviation pattern y' registered in advance in the deviation pattern database 18e. k and the abnormal part of equipment A, a deviation pattern y similar to deviation pattern y' is found. k The data table shown in FIG. 7( b) records deviation patterns obtained when an abnormality occurs in the equipment and organizes them in association with the abnormal part. This data table is created in advance and stored in the deviation pattern database 18e by the normal distribution determination and recording unit 14. The data table can also be configured to identify the abnormal part based on deviation patterns that take into account m types of operating conditions in addition to classification into k types of operating sections and l types of frequency bands. Therefore, the abnormal part identification unit 17 may identify the abnormal part of equipment A by taking into account m types of operating conditions.

[0040] The abnormal part specifying unit 17 uses the Euclidean distance to determine the deviation pattern y k Calculate the similarity of the deviation pattern y′ to the deviation pattern y′, and find the similar deviation pattern y based on the calculated similarity. k The similarity shown in formula (1) is calculated by searching for the deviation pattern y′. k The more similar the deviation pattern y is to , the closer the value is to 1. k For example, in the example shown in FIGS. 7A and 7B, deviation pattern y′ and deviation pattern y generated in the process of step S14 are read out. 2 Since these patterns match, the abnormal part specifying unit 17 determines that the deviation pattern y 2 The part B corresponding to the part B is identified as the abnormal part of the equipment A. This completes the process of step S15, and the series of abnormal part identification processes ends.

[0041]

[0042] As is clear from the above description, the acoustic diagnostic device 1 according to one embodiment of the present invention comprises an operation sound collection unit 11 that collects operation sounds of equipment A, an operation section operation sound extraction unit 12 that collects images showing the operating state of equipment A using an image sensor 21, detects predetermined operation sections of equipment A using the collected images of equipment A, and extracts operation sounds for each operation section detected from the operation sounds collected by the operation sound collection unit 11, a sound pressure level acquisition unit 15 that acquires sound pressure levels for each predetermined frequency band for each operation section for the operation sounds extracted by the operation section operation sound extraction unit 12, a deviation degree calculation unit 16 that calculates the degree of deviation from the sound pressure level during normal operation of the same equipment, operation section, and frequency band for the sound pressure levels for each operation section and each frequency band acquired by the sound pressure level acquisition unit 15, and an abnormal part identification unit 17 that identifies an abnormal part of the equipment based on the degree of deviation calculated by the deviation degree calculation unit 16. With this configuration, abnormal parts of equipment A are identified based on the sound pressure levels for each operating section and each frequency band of equipment A, allowing for accurate identification of abnormal parts of equipment A. Furthermore, since the operating section of equipment A is detected using the image sensor 21, the acoustic diagnostic device 1 can operate independently. This also enables swift adjustment, repair, or replacement of abnormal parts, making it possible to prevent deterioration or defects in the quality of manufactured products. This can also contribute to improving product yield.

[0043] The acoustic diagnostic device 1 according to one embodiment of the present invention further includes a normal state sound pressure level acquisition unit 13 that acquires sound pressure levels for predetermined frequency bands for each operation interval for operation sounds during normal operation of the equipment A extracted by the operation interval operation sound extraction unit 12, and a normal distribution determination and recording unit 14 that classifies the sound pressure levels acquired by the normal state sound pressure level acquisition unit 13 for each equipment, each operation interval, and each frequency band, records them as sound pressure levels during normal operation of the equipment A, and records a table for each operation interval in which deviation patterns obtained when an abnormality occurs in the equipment A for each frequency band are associated with abnormal parts. With this configuration, for operation sounds during normal operation of the equipment A, sound pressure levels for each equipment, each operation interval, and each frequency band, and a table in which deviation patterns obtained when an abnormality occurs and abnormal parts are associated are recorded. As a result, even if there is a change in the operating sound during normal operation, new recording processing of the operating sound during normal operation is performed, and the abnormal part of the equipment A is identified based on the sound pressure level for each operating section and each frequency band using the new operating sound data during normal operation as a reference, thereby making it possible to accurately identify the abnormal part of the equipment A. Furthermore, since the operating section of the equipment A is detected using the image sensor 21, the acoustic diagnostic device 1 can operate independently.

[0044] Furthermore, the acoustic diagnostic device 1 according to one embodiment of the present invention identifies an abnormal portion of equipment A while taking into account the operating conditions of equipment A, thereby enabling accurate identification of the abnormal portion of equipment A. Generally, in steelmaking equipment such as press equipment and rolling equipment, even during normal operation, operating conditions such as the amount of rolling and materials change, resulting in corresponding variations in the loudness and frequency of the operating sounds. As a result, the normal range of the loudness and frequency of the operating sounds widens, and the loudness and frequency of abnormal sounds are buried within the normal range, resulting in the inability to accurately detect an abnormality in the equipment based solely on the loudness and frequency of the operating sounds. In contrast, the acoustic diagnostic device 1 according to one embodiment of the present invention identifies an abnormal portion of equipment A while taking into account the operating conditions of equipment A, thereby enabling accurate identification of the abnormal portion of equipment A.

[0045] In this example, abnormal parts of the press equipment were identified. By installing acoustic sensors around the press equipment, the spectrum of the impact sound could be acquired the moment a press load was applied, and the operating sound of the conveyor could be acquired while it was in operation after pressing was completed. Furthermore, by installing cameras on the entry and exit sides of the press equipment, images of the slab entering the press equipment and the slab leaving the press equipment, as shown in Figures 8(a) and 8(b), could be collected. In this example, the images shown in Figures 8(a) and 8(b) were input into an artificial intelligence model to detect the time when the workpiece entered the press equipment and the time when the workpiece left the press equipment. Then, by taking into account the operating conditions of the press equipment, the range of the frequency distribution of the sound pressure level during normal operation of the press equipment was narrowed, and abnormal parts of the press equipment were identified. FIG. 9( a) shows the frequency distribution of sound pressure levels when the operating conditions of the press equipment are limited (average value 87.6 dB, standard deviation σ 2.19), and FIG. 9( b) shows the frequency distribution of sound pressure levels when the operating conditions of the press equipment are not limited (average value 85.7 dB, standard deviation σ 1.6).

[0046] Next, assuming a die abnormality, the deviation degree of the press equipment when the sound pressure level was 91 dB was calculated. The following formula (2) was used to calculate the deviation degree of the press equipment. As a result, when the operating conditions were not limited, the deviation degree of the press equipment when the sound pressure level was 91 dB was calculated to be 1.5, and when the operating conditions were limited, the deviation degree of the press equipment when the sound pressure level was 91 dB was calculated to be 3.3. Therefore, if the deviation threshold for determining that an abnormality has occurred in the press equipment is set to 3, it is possible to accurately determine whether an abnormality has occurred in the press equipment. Furthermore, by searching for a deviation pattern similar to the calculated deviation pattern from deviation patterns registered in advance, it was possible to accurately determine that the abnormal part of the equipment was the die.

[0047]

[0048] Although the present invention has been described above as an embodiment, the present invention is not limited to the descriptions and drawings that form part of the disclosure of the present invention. In other words, other embodiments, examples, and operational techniques that can be made by those skilled in the art based on the present invention are all included in the scope of the present invention.

[0049] According to the present invention, it is possible to provide an acoustic diagnostic device and an acoustic diagnostic method that can accurately identify an abnormal portion of equipment.

[0050] REFERENCE SIGNS LIST 1 Acoustic diagnostic device 11 Operation sound collection unit 12 Operation section operation sound extraction unit 13 Normal state sound pressure level acquisition unit 14 Normal distribution determination and recording unit 15 Sound pressure level acquisition unit 16 Deviation degree calculation unit 17 Abnormal part identification unit 18 Database unit 18a Extraction condition database 18b Frequency band database 18c Normal distribution database 18d Operation condition database 18e Deviation pattern database 20 Acoustic sensor A Equipment to be diagnosed

Claims

1. An acoustic diagnostic device that identifies abnormal parts of equipment using acoustic signals generated during operation of the equipment, comprising: an operation sound collection unit that collects operation sounds of the equipment; an operation section operation sound extraction unit that collects images showing the operating state of the equipment, detects predetermined operation sections of the equipment using the collected images, and extracts the operation sounds for each operation section detected from the operation sounds collected by the operation sound collection unit; a sound pressure level acquisition unit that acquires sound pressure levels for each frequency band predetermined for each operation section for the operation sounds extracted by the operation section operation sound extraction unit; a deviation degree calculation unit that calculates a deviation degree from a sound pressure level during normal operation of the same equipment, operation section, and frequency band for the sound pressure levels for each operation section and for each frequency band acquired by the sound pressure level acquisition unit; and an abnormal part identification unit that identifies abnormal parts of the equipment based on the deviation degree calculated by the deviation degree calculation unit.

2. The acoustic diagnostic device of claim 1, wherein the operation section operation sound extraction unit detects a predetermined operation section of the equipment using the image showing the operation state of the equipment by inputting an image showing the operation state of the equipment acquired using an image sensor into an artificial intelligence model having an image showing the operation state of the equipment as input data and an operation state of the equipment shown in the image as output data.

3. The acoustic diagnostic device according to claim 1 or 2, wherein the sound pressure level acquisition unit classifies each operation sound based on predetermined operating conditions of the equipment and acquires a sound pressure level, and the deviation calculation unit calculates a deviation from a sound pressure level during normal operation of the same equipment, operating section, frequency band, and operating condition for the sound pressure level acquired by the sound pressure level acquisition unit for each operating section, frequency band, and operating condition.

4. An acoustic diagnostic device as described in claim 3, further comprising a database section for storing conditions for extracting the operating sounds of the equipment and data required for identifying an abnormal part of the equipment, wherein the abnormal part identification section identifies an abnormal part of the equipment based on a pattern of deviation degrees that are elements of the deviation degrees in each frequency band obtained for each operating section stored in the database section.

5. An acoustic diagnostic device as claimed in any one of claims 1 to 4, comprising: a normal sound pressure level acquisition unit that acquires sound pressure levels for each predetermined frequency band for each operation section for the operation sound extracted by the operation section operation sound extraction unit; and a normal distribution determination and recording unit that classifies the sound pressure levels acquired by the normal sound pressure level acquisition unit for each equipment, each operation section, and each frequency band, and records them as sound pressure levels during normal operation of the equipment.

6. The acoustic diagnostic device according to claim 5, wherein the normal distribution determination and recording unit records, for each piece of equipment, a table including information showing the relationship between a pattern of deviations of sound pressure levels from the sound pressure level during normal operation in each frequency band determined for each operating section and abnormal parts of the equipment.

7. An acoustic diagnostic device as described in claim 6, wherein the normal distribution determination and recording unit records a table including information indicating the operating section, the pattern of the deviation degree, and the relationship between the operating conditions of the equipment and the abnormal parts of the equipment.

8. An acoustic diagnosis method for identifying an abnormal part of equipment using an acoustic signal generated during operation of the equipment, comprising: an operation sound collection step of collecting operation sounds of the equipment; an operation section operation sound extraction step of collecting images showing the operating state of the equipment, detecting predetermined operation sections of the equipment using the collected images, and extracting the operation sounds for each operation section detected from the operation sounds collected in the operation sound collection step; a sound pressure level acquisition step of acquiring a sound pressure level for each frequency band predetermined for each operation section for the operation sounds extracted in the operation section operation sound extraction step; a deviation calculation step of calculating a degree of deviation from a sound pressure level during normal operation of the same equipment, operation section, and frequency band for the sound pressure levels for each operation section and for each frequency band acquired in the sound pressure level acquisition step; and an abnormal part identification step of identifying an abnormal part of the equipment based on the degree of deviation calculated in the deviation calculation step.

9. An acoustic diagnostic method as described in claim 8, comprising: a normal sound pressure level acquisition step of acquiring a sound pressure level for each predetermined frequency band for each operation section for the operation sound extracted in the operation section operation sound extraction step; and a normal distribution determination and recording step of classifying the sound pressure levels acquired in the normal sound pressure level acquisition step for each equipment, each operation section, and each frequency band, and recording them as sound pressure levels during normal operation of the equipment.

10. An acoustic diagnostic method as described in claim 8 or 9, wherein the operation interval operation sound extraction step includes a step of inputting an image showing the operation state of the equipment acquired using an image sensor into an artificial intelligence model having an image showing the operation state of the equipment as input data and an operation state of the equipment shown in the image as output data, thereby detecting a predetermined operation interval of the equipment using the image showing the operation state of the equipment.

Citation Information

Patent Citations

  • Diagnostic apparatus

    JP2008164490A

  • State estimation device, method, and program

    WO2019017345A1

  • Diagnostic device, equipment comprising same, and diagnostic method

    WO2020157818A1

  • Vibration monitoring system suitable for power equipment

    CN115200699A

  • Device for diagnosing abnormal operation noise of equipment

    JP2013019741A

Cited By

  • Acoustic diagnostic device and acoustic diagnostic method

    JP7827233B1

  • Acoustic diagnostic device and acoustic diagnosis method

    WO2026083670A1